Investigation of the Properties ofX-ray Pulsars During Outbursts

A Thesis

submitted for the Award of Ph.D. degree of

MOHANLAL SUKHADIA UNIVERSITY

in the

Faculty of Science

By

Prahlad Rameya Epili

Under the supervision of

Dr. Sachindranatha NaikAssociate Professor

Astronomy and Astrophysics DivisionPhysical Research Laboratory, Ahmedabad, India

DEPARTMENT OF PHYSICS

FACULTY OF SCIENCE

MOHANLAL SUKHADIA UNIVERSITY

UDAIPUR (RAJ)

Year of submission: 2018

DECLARATION

I, Mr. Prahlad Rameya Epili, S/o Shri. Rameya Narayan Epili and Smt. Bhanu-

mati Epili, resident of 209, PRL Thaltej Hostel, Ahmedabad 380054, hereby declare

that the research work incorporated in the present thesis entitled, “Investigation of

the Properties of X-ray Pulsars During Outbursts” is my work and is original.

This work (in part or in full) has not been submitted to any University for the award

of a degree or a diploma.

I have properly acknowledged the material collected from the secondary sources

wherever required and I have run my entire thesis on the antiplagiarism software

namely, iThenticate Tool.

I solely own the responsibility for the originality of the entire content.

Date:

Prahlad Rameya Epili

(Author)

CERTIFICATE

I feel great pleasure in certifying that the thesis entitled, “Investigation of the

Properties of X-ray Pulsars During Outbursts” embodies a record of the results

of investigations carried out by Mr. Prahlad Rameya Epili under my guidance. He

has completed the following requirements as per Ph.D regulations of the University.

(a) Course work as per the university rules.

(b) Residential requirements of the university.

(c) Regularly submitted six monthly progress reports.

(d) Presented his work in the departmental committee.

(e) Published minimum of one research paper in a referred research journal.

I recommend the submission of thesis.

Date:

Dr. Sachindranatha Naik

Associate Professor,

Astronomy and Astrophysics Division,

Physical Research Laboratory,

Ahmedabad - 380 009.

Countersigned by

Head of the Department

Dedicated to

my loving parents

Acknowledgements

The last five years of my journey venturing upon the audacious task of completing this thesis

work would not have possible without the active support of many people, whose paths I have

crossed for the first time and instantly became friends with and some were life long source of

inspiration. Humbly, I acknowledge all these wonderful people whose contribution is not

less than mine in bringing up this thesis to its present form.

I would like to thank my thesis supervisor Dr. Sachindra Naik, for accepting me as his

PhD student and being my persistent supporter and mentor throughout my PhD journey. It

turns out that in all the matters related to my academics, working on a research problem or

be it be in making a conclusion of a particular work, his kind remarks and suggestions were

invaluable and have always brought much ease and fruitful results. Truly speaking, without

his constant support and encouragement, it would have been impossible for me to let this

manuscript saw the light of the day. I would be eternally thankful to him for giving a patient

ear to all my silly questions pertaining to X-ray Astronomy.

I am thankful to my thesis Doctoral Student Committee (DSC) members, Prof. Kiran

S. Baliyan and Dr. Santosh Vadawale for keeping an eye on my progress through all these

years. As and when required, they have provided necessary remarks related to my thesis

work. It also goes without saying that, they have meticulously gone through my thesis draft

and pointed out few places where further improvement is needed. During the course of PhD

work, I am greatly benefited by presenting my progress in thesis work time to time in front

of academic committee. I am thankful to Academic Committee Chairman of PRL and other

members who took a keen interest in my progress. Its an honour to express my thankfulness

to present Director of PRL, Dr. Anil Bhardwaj who has kindly allowed me to present my

thesis results in the workshops and conferences held at national at international level. I have

highly benefited by attending these events.

I have greatly benefited from the innumerable academic interactions with faculty and

research staff of Astronomy & Astrophysics division (A&A Div.) of PRL. I am especially

viii

thankful to Prof. Abhijit Chakrabarty & Prof. Kiran S. Baliyan, respectively the present and

past chairmen of A&A Division who have kindly allowed me to present my research work at

international meetings and conferences and also allocated the necessary travel grants to

attend these meetings. I would like to offer my special thanks to Prof. N. M. Ashok, Prof. U.C.

Joshi, Prof. P. Janardhan, Dr. Sashikiran Ganesh, Dr. Aveek Sarkar, Dr. Mudit Srivastava,

Dr. Veeresh Singh, Dr. Lokesh Dewangan, Dr. Arvind S. Rajpurohit, Dr. Vishal Joshi for

their encouragements and support at various stages of my PhD tenure. Their critical remarks

during my area seminars have strengthened my presentation skills and also helped me to

think clearly. Among the other staff of A&A Division, I thank immensely to Mr. Vaibhav

Dixit, Mr. Mithun Neelkandan PS, Mr. Neeraj Tiwari, Mr. Kapil Kumar, Mr. J S S V Prasad

Neelam, Mr. Rishikesh Sharma, Ms. Deekshya Roy Sarkar, Mr. Mohan Lal, Ms. Ankita Patel,

Mr. Vishal M Shah, Mr. P. S. Patwal for helping me out on a number of occasions related to

scientific and administrative matters.

Although most of my thesis work dealt with the X-ray observations of Be/X-ray binary

pulsars, but I was also keenly interested in taking IR-observations of the companion Be stars.

For this, I was actively participating in the IR-observation campaign on Be stars, which is led

by my mentor Dr. Sachindra Naik. I owe a lot to the Mt. Abu IR-Observatory (MIRO) staff

for their kind hospitality and support during my each visit for the IR-observation campaign.

Especially, Mr. J. K. Jain, Mr. Nafees Ahmed, Mr. G. S. Rajpurohit, Mr. R. R. Shah, Mr. S.

N. Mathur were very cordial and approachable during times when the telescope operation

have to be changed from taking photometric to spectroscopic mode, during the observations.

Their vast experience of handling and maintenance of 1.2 Optical/IR telescope at MIRO

have helped me in taking observations smoothly. I will always remember their unflinching

support and allowing me to carry out IR observations throughout the night without worrying

about the intricate details of telescope handling and maneuver. I also thank Mr. Vivek Vicky

Unyal and his staff at PRL Hill-View guest, Mt. Abu for making my stay comfortable there.

I take this opportunity to thank Gaurava Jaisawal (now at Technical University of

Denmark) with whom I have been closely associated for the last 3 years of my PhD tenure.

In many ways, Gaurava had been instrumental in all of my academic pursuits during the

PhD tenure. I am highly benefited from his quick remarks and judgments on my writing and

presentation skills. We had the opportunity share common thesis supervisor and common

lab space during all those years. We had many stimulating discussions related to X-ray

astronomy research and argued over many basic topics related to our thesis works and

ix

properties of X-ray pulsars, in general. I should admire the fact that his ways of putting facts

and logic in congruent with proposed topic of discussion would yield in a clear picturesque

description of the subject. My other labmates presently Dr. Main Pal, Shivangi, Neeraj and

Dr. Neelkanth in past, with whom I had the benefit of chit-chatting on a number of topics

during lunch and coffee breaks and have always made me felt homely. There were funny

instances where we would burst into laughter together. I would cherish these memories.

I also thank my other senior colleagues at PRL with whom I had the opportunity to

discuss freely about my other interests and seek support in times of need. Its a boon to

have wonderful seniors like Tanmoy, Arko, Priyanka, Sushant, Sunil, Girish Chakrabarty,

Amerandra, Monojit, Naveen Negi, Diptiranjan, Deepak, Bivin, Lalit with whom I had the

thrilling discussions on many interesting topics during weekends. Their cordial support and

friendly gesture during my initial years at PRL-Thaltej students hostel has entrusted in me

the confidence to carry out the present thesis work.

It would be nearly impossible for me to deny the unwavering support and love that I have

received from my fellow batch-mates: Satish, Kuldeep, Navpreet, Rukmani, Rupa, Kumar

V., Jabir, Ali, Chandan during the last five years. We have made it a point to celebrate each

ones little success an opportunity to call for celebration and party every time. I will cherish

these moments for a long time in my life and wish I could be there with them in every ups

and downs in their life.

It’s always a pleasure to see ones predecessors venturing upon the similar paths one has

came across and sharing their pleasant and triumphant occasions. It brings a jolly mood and

made my day several times after friendly chat round the dinner tables at PRL-Thaltej canteen

with fellow companions like Archita, Aarty, Shivangi, Richa, Subir, Kaustav, Varun, Shefali,

Needhi, Rahul, Akanksha. I also thank some of my colleagues from PRL (Navrangpura

Campus), Arvind Mishra, Ashish with whom occasionally I had discussed topics of common

interest over the phone or during my infrequent visits to the campus. Thaltej Campus of PRL

had been my second home cum place of research during the last five years. Here, I had the

benefit of cherishing friendship with Chandan Gupta, Nisha, Priyank, Sandeep, Sushree,

Ayan, Abdur, Deepika, Harish and enjoyed celebrating festivals like holi, garba, new year

celebrations (having horror themes) with great fervor and enthusiasm.

In the matters of administrative assistance, I have always looked forward to Shri Anand

D. Mehta, Bhagirathi Bhai and Shri Pinakin Shikari and have got their kind support even at

the eleventh hour. I am highly thankful for their timely support. I also thank the staff of PRL

x

library Dr. Md. Nurul Alam and Dr. Nishtha Anilkumar for allowing me to issue a number

of reference books at a time and keep with me when I needed it mostly. I thank profusely

for this. The variety of research journals subscriptions and manuscripts available at PRL

library have always caught my attention to browse through them in times when I was visiting

the library. Its a pleasure to acknowledge all the library staff.

During my undergraduate years (2006-2009), I was fortunate to have associated with

wonderful lecturers like Dr. Atul Mody, Ms. Sulekha Gopinath, Ms. Hemlatha Deshpande,

Dr. Ms. Anita Kanwar, Ms. Latika Menon in Physics Department of V.E.S. College, Chembur

(Mumbai). Their methods of teaching physics and make it easily grasp the concepts had

pondered me a lot of times in terms of exciting world of physics out there. Particularly, Dr.

Atul Mody had been instrumental in bringing to my knowledge the various sub-fields of

physics research. From his friendly nature and tireless efforts of conducting all those physics

problem solving sessions during summer of 2009 to bunch of physics undergraduates, I have

greatly benefited and uphold a greater degree of admiration for him. Apart from academics,

I find him an easily approachable person in times of need and support. I thank him from the

bottom of my heart for all the encouragement he bestowed upon me during my PhD tenure.

To my lovely sisters Mamata & Pramila, I would remain highly indebted for their

patience in all these years and letting me finish this thesis in due course. If there are any

two people to whom I would remain highly indebted beyond the realms of this material life,

they are my parents, Shri Rameya Epili and Smt. Bhanumati Epili. They have been the

whole reason in all of my endeavours. Their everlasting love and support for the cause of my

education, happiness & well being made me ponder what better gift I would ask to almighty.

Finally I am thankful to all the people that I have interacted so far, whose direct and indirect

contribution have led me to the completion of this thesis.

Prahlad

Abstract

The primary aim of this thesis is to investigate the properties of Be/X-rays binary

pulsars during different type of outbursts. Accretion powered X-ray pulsars are

among the luminous X-ray sources in the Galaxy. The X-ray luminosity of these

sources varies in Lx ∼ 1034 −1038 ergss−1 range. The compact star in the system

happens to be a spinning neutron star endowed with strong magnetic field (1012 G).

The spin period of these neutron stars is in the range of sub-seconds to thousands of

seconds. The primary source of X-ray emission from these objects is due to accretion

of matter from the companion star. The accreted matter follows the strong magnetic

field lines and channels onto the poles of the neutron star. Near the polar regions of

the neutron star, the accreting matter is being decelerated through a radiative shock in

the accretion column before it settles on the surface. At base of the accretion column,

the in-falling matter with terminal speeds of ∼ 0.3c is abruptly stopped. The poles

of the neutron star get heated up due to the transfer of energy of the accreted matter

and emit beams of radiation in X-ray range. The interaction of this radiation with

the incoming plasma in the accretion column results in broad-band X-ray emission

due to complex physical processes.

In Be/X-ray binary pulsars, the compact object (neutron star) accrets matter

from the evolving circumstellar disk of the companion Be star at periastron passage.

This coincides with the enhancement in X-ray emission from the neutrons tar. The

X-ray luminosity of the pulsar hence increases by an order of magnitude (i.e ∼

1036 −1037ergss−1) during these outbursts. However, there are rare and irregular

events representing significant increase in the luminosity of the neutron star due to

abrupt mass accretion from the Be star disk. The X-ray luminosity of the pulsar

during these rare and unpredictable events reaches as high as ≥ 1038ergss−1. These

xii

regular and irregular events leading to enhancement in X-ray emission from the

pulsar are known as Type-I and Type-II outbursts. This kind of X-ray variabilities in

Be/X-ray binaries make them unique among other binary sources. Secondly, the X-

ray observations during these events make it feasible to explore the column emission

and variety of pulsar related phenomena. As these events cover a wide luminosity

range (i.e 1036 −1038 ergss−1), it provides an unique opportunity to investigate the

beamed emission geometry of the pulsar through temporal and spectral studies of

these sources. Although the rare Type-II outbursts of a few BeXBs have been studied

in the past using X-ray observations, cause of such events and their origin remains

inconclusive till date.

Observational study of these pulsars during outbursts gives us an opportunity to

understand the dynamics of X-ray emission from the poles and accretion column. It

also helps us in estimating the pulsar magnetic field, one of the fundamental property

of the neutron star. In this thesis, I have used observations spanned over more than

a decade, comprising many outbursts of Be/X-ray binary pulsar EXO 2030+375

to understand the emission mechanism and properties during different types of

outbursts. The timing and spectral studies of this pulsar revealed several key features

of the accretion column emission that were remained hitherto unknown. In particular,

investigation of patterns of pulse profiles of this pulsar in a broad luminosity range (i.e

Lx ∼ 1036−1038 ergss−1) revealed the structural changes in the beamed emission. In

addition to this, the robustness of the dependence of the pulse profile on luminosity,

irrespective of the phases and types of X-ray outbursts are clearly reflected across

a wide range of pulsar luminosity. Moreover, the spectral state transition near the

critical luminosity regime of the pulsar is also observed from the variable photon

index and other spectral parameters. This also showed that the beamed emission

from the pulsar changes its geometry from a pencil beam at lower luminosity to a

mixture of pencil and fan beam emission at the transitional luminosity. As the peak

luminosity during Type-II events reaches the Eddington limit, the structure of column

emission is dominated by the fan beam geometry. At low luminosity of the pulsar, we

observed the evolution of a narrow peculiar dip in the pulse profiles. Subsequently, it

was found out that the self obscuring accretion column was resulting in such narrow

xiii

dip like features in the pulse profile. These changes in the column emission is also

studied by using a physical model in addition to several continuum models. The

long term X-ray variability studies of EXO 2030+375 provided this compelling

evidence that the structure and geometry of column emission from the pulsar do not

depend upon the Type-I or Type-II outburst activity. Rather it solely depends on

the mass accretion rate onto the pulsar. The physical model developed by Becker &

Wolff (2007) (i.e BW model) by considering the thermal & bulk Comptonization

in the X-ray pulsar accretion column has been used for the first time to explain

the broad-band X-ray spectra of EXO 2030+375. Through this model, we found

the changes in physical parameters of the pulsar with varying mass accretion rate

across outbursts. Luminosity changes in the parameters like accretion column radius,

diffusion rate, ratio of thermal to bulk Comptonization, plasma temperature in the

accretion column have illustrated the importance of critical luminosity in shaping

the pulsar emission geometry. The magnetic field of this pulsar could be constrained

with the BW model which otherwise remained uncertain due to ambiguous detection

of Cyclotron Resonance Scattering Features (CRSFs) in the pulsar spectra.

Apart from the extensive studies of EXO 2030+375, I have also investigated the

properties of another Be/X-ray binary pulsar KS 1947+300 during a rare Type-II

outburst in October-November, 2013 by using Suzaku observations at two epochs.

From this, we have found some of the interesting properties of pulsar emission as

it emerges from the dense accreting plasma. The energy resolved pulse profiles

of KS 1947+300 revealed the X-ray pulsations from pulsar upto energies as high

as 150 keV. And more interestingly, the pulse profiles at soft X-rays (i.e below

10 keV) were smooth and single peaked. With increase in energy, however, there

were sharp dips at certain pulse phases seen to be evolving upto energies ≤ 55 keV .

To understand the cause of such absorption dips, we have carried out phase-averaged

and phase-resolved spectral studies of the pulsar. From these studies, we found that

the soft X-ray excess component to be pulsating in phase with the pulsar continuum

emission. This revealed the regions close to the pulsar i.e. accretion column could

be the probable sites for its origin. This also explained the absence of intrinsic dips

in the soft X-ray pulse profiles which otherwise were clear and found evolving in

xiv

hard X-ray pulse profiles. From the strong spin-up of the pulsar, as observed during

the giant outburst, we estimated the dipole magnetic moment of the pulsar by using

theory of quasi-spherical settling accretion. The pulsar magnetic field estimated

from this was found to be in agreement with the independent estimates from pulsar

cyclotron lines.

Keywords: X-ray binaries, Accretion, Pulse profiles, Accretion powered X-ray

pulsars, Spectroscopy, Pulse phase resolved spectroscopy, Magnetic fields, Critical

luminosity

Contents

List of Figures xix

List of Tables xxi

1 Introduction 1

1.1 X-ray Astronomy: The early years . . . . . . . . . . . . . . . . . . 1

1.2 X-ray Binaries and their Classification . . . . . . . . . . . . . . . . 2

1.2.1 Low Mass X-ray Binaries . . . . . . . . . . . . . . . . . . 2

1.2.2 High Mass X-ray Binaries . . . . . . . . . . . . . . . . . . 3

1.3 Accretion in X-ray binaries . . . . . . . . . . . . . . . . . . . . . . 7

1.3.1 Accretion onto neutron star X-ray binaries . . . . . . . . . . 8

1.4 Characteristics Properties of Neutron Stars . . . . . . . . . . . . . . 11

1.5 Accretion powered X-ray Pulsars . . . . . . . . . . . . . . . . . . . 17

1.5.1 Critical & Eddington Luminosity . . . . . . . . . . . . . . 20

1.6 Emission mechanism in accreting X-ray pulsars . . . . . . . . . . . 21

1.6.1 Accretion regimes in X-ray pulsars . . . . . . . . . . . . . 22

1.6.2 Phenomenological Spectral Models . . . . . . . . . . . . . 24

1.7 Motivation for present thesis . . . . . . . . . . . . . . . . . . . . . 26

1.8 Thesis Outline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28

2 X-Ray Observations & Data Analysis 31

2.1 X-ray Observations . . . . . . . . . . . . . . . . . . . . . . . . . . 31

2.2 X-ray Telescope . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

2.3 X-ray Detectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

2.4 Rossi X-ray Timing Explorer (1995 - 2012) . . . . . . . . . . . . . 34

xvi Contents

2.4.1 Proportional Counter Array (PCA) . . . . . . . . . . . . . . 35

2.4.2 High Energy X-ray Timing Experiment (HEXTE) . . . . . . 36

2.5 Suzaku . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

2.5.1 X-ray Imaging Spectrometers (XIS) . . . . . . . . . . . . . 38

2.5.2 Hard X-ray Detector (HXD) . . . . . . . . . . . . . . . . . 40

2.6 NuSTAR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

2.7 Data Reduction Techniques, Software and

Analysis Procedures . . . . . . . . . . . . . . . . . . . . . . . . . 43

3 2013 giant X-ray outburst of Be/X-ray binary pulsar KS 1947+300 49

3.1 Be/X-ray binary pulsar KS 1947+300 . . . . . . . . . . . . . . . . 49

3.2 Data Reduction and Analysis . . . . . . . . . . . . . . . . . . . . . 52

3.3 Pulse profiles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54

3.4 Pulse-Phase Averaged Spectroscopy . . . . . . . . . . . . . . . . . 57

3.5 Pulse-phase Resolved Spectroscopy . . . . . . . . . . . . . . . . . 60

3.6 Results & Discussion . . . . . . . . . . . . . . . . . . . . . . . . . 65

3.7 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69

4 Timing and spectral studies of EXO 2030+375 during X-ray outbursts 71

4.1 Be/X-ray binary pulsar EXO 2030+375 . . . . . . . . . . . . . . . 71

4.2 Observations and Analysis of EXO 2030+375 . . . . . . . . . . . . 74

4.2.1 Luminosity dependent pulse profiles . . . . . . . . . . . . . 76

4.3 Phase Averaged Spectroscopy . . . . . . . . . . . . . . . . . . . . 82

4.4 Results & Discussion . . . . . . . . . . . . . . . . . . . . . . . . . 88

5 Accretion column emission in EXO 2030+375: a physical perspective 95

5.1 Physical Spectral Model . . . . . . . . . . . . . . . . . . . . . . . 96

5.1.1 Thermal and Bulk Comptonization model . . . . . . . . . . 97

5.1.2 BW model explaining the spectra of EXO 2030+375 . . . . 99

5.2 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104

6 Summary and Future Scope 107

6.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108

Contents xvii

6.2 Future Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112

References 113

List of Publications . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127

Publications attached with Thesis . . . . . . . . . . . . . . . . . . . . 129

List of Figures

1.1 Accretion in BeXBs . . . . . . . . . . . . . . . . . . . . . . . . . . 6

1.2 Roche-lobe Potential . . . . . . . . . . . . . . . . . . . . . . . . . 10

1.3 stellar-wind accretion . . . . . . . . . . . . . . . . . . . . . . . . . 12

1.4 Representation of neutron star populations . . . . . . . . . . . . . . 15

1.5 X-ray pulsar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

1.6 pencil- and fan beam emission geometry . . . . . . . . . . . . . . . 23

2.1 Wolter Type-I optics . . . . . . . . . . . . . . . . . . . . . . . . . 33

2.2 Rossi X-ray Timing Explorer . . . . . . . . . . . . . . . . . . . . . 35

2.3 Suzaku X-ray observatory . . . . . . . . . . . . . . . . . . . . . . 38

2.4 Suzaku/XIS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

2.5 NuSTAR X-ray observatory . . . . . . . . . . . . . . . . . . . . . 42

3.1 Swift/BAT lightcurve of KS 1947+300 during 2013 giant outburst . 53

3.2 Pulse profiles of KS 1947+300 during 2013 Type II outburst . . . . 55

3.3 Energy resolved pulse profiles of KS 1947+300 during the first

Suzaku observation . . . . . . . . . . . . . . . . . . . . . . . . . . 56

3.4 Energy resolved pulse profiles of KS 1947+300 during the second

Suzaku observation . . . . . . . . . . . . . . . . . . . . . . . . . . 56

3.5 Broadband energy spectrum of KS 1947+300 during 2013 Giant

outburst as seen from first Suzaku observation . . . . . . . . . . . . 59

3.6 Broadband energy spectrum of KS 1947+300 during 2013 Giant

outburst as seen from second Suzaku observation . . . . . . . . . . 60

xx List of Figures

3.7 Change in spectral parameters over pulse phases of the pulsar during

Obs I . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62

3.8 Change in spectral parameters over pulse phases of the pulsar during

Obs II . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63

3.9 Correlation between various spectral parameters during Obs I . . . . 64

3.10 Correlation between various spectral parameters during Obs II . . . 64

4.1 The 2-60 keV pulse profiles of EXO 2030+375 in wide range of

pulsar luminosity. . . . . . . . . . . . . . . . . . . . . . . . . . . . 77

4.2 Comparison of the pulse profiles of EXO 2030+375 at similar lumi-

nosity levels during Type I and Type II X-ray outburst . . . . . . . . 78

4.3 Pulse profiles of EXO 2030+375 at similar luminosity during the

rising and declining phases of the 2006 giant outburst. . . . . . . . . 80

4.4 Evolution of a narrow absorption dip at low X-ray luminosity in the

pulse profiles of EXO 2030+375 . . . . . . . . . . . . . . . . . . . 81

4.5 Phase-averaged spectra of EXO 2030+375, obtained from the RXTE

observations of the pulsar at different luminosity levels . . . . . . . 83

4.6 3-100 keV spectra of EXO 2030+375 obtained from the RXTE ob-

servation at the peak of 2006 giant outburst . . . . . . . . . . . . . 84

4.7 Spectral parameters obtained from fitting of RXTE observations of

EXO 2030+375 during Type I & Type II outbursts with 3-30 keV

luminosity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87

5.1 Schematic representation of matter accretion onto the pole of a

neutron star . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98

5.2 Phase-averaged energy spectra of EXO 2030+375 at three distinct

luminosities obtained during Type I and Type II X-ray outbursts . . 101

5.3 Spectral parameters obtained from the fitting of phase averaged

spectra of EXO 2030+375 with BW model at different luminosities 103

List of Tables

2.1 Characteristics of X-ray Detectors on-board various X-ray observato-

ries (such as Suzaku , NuSTAR , RXTE ). Data from these instruments

has been studied in the present thesis. . . . . . . . . . . . . . . . . 44

3.1 Log for the Suzaku observations of KS 1947+300 during 2013 Type-

II outburst. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52

3.2 Best-fit parameters of phase-averaged spectra for KS 1947+300 from

Suzaku observations . . . . . . . . . . . . . . . . . . . . . . . . . 61

4.1 Log of RXTE/PCA observations of the pulsar EXO 2030+375 during

Type I and Type II outbursts. . . . . . . . . . . . . . . . . . . . . . 75

4.2 Best-fitting spectral parameters with 1σ errors obtained from RXTE/PCA

observations of EXO 2030+375 at six different luminosities. The

best-fit model consists of a partial covering high-energy cutoff power-

law model with a Gaussian component. . . . . . . . . . . . . . . . 86

5.1 BW model parameters obtained from spectral fitting of EXO 2030+375

data obtained from RXTE /PCA and HEXTE observations during

different luminosity epochs of the pulsar . . . . . . . . . . . . . . . 100

References

Abdo, A. A., Ajello, M., Allafort, A., Baldini, L., Ballet, J., Barbiellini, G., Baring,

M. G., Bastieri, D., Belfiore, A., Bellazzini, R., and et al. (2013). The Second

Fermi Large Area Telescope Catalog of Gamma-Ray Pulsars. Astrophys. J.

Supplement., 208:17.

Arnaud, K. A. (1996). XSPEC: The First Ten Years. In Jacoby, G. H. and Barnes,

J., editors, Astronomical Data Analysis Software and Systems V, volume 101 of

Astronomical Society of the Pacific Conference Series, page 17.

Basko, M. M. and Sunyaev, R. A. (1976). The limiting luminosity of accreting

neutron stars with magnetic fields. Mon. Not. Roy. Astron. Soc., 175:395–417.

Baykal, A., Kızıloglu, Ü., Kızıloglu, N., Balman, S., and Inam, S. Ç. (2005). X-ray

outburst of 4U 0115+634 and ROTSE observations of its optical counterpart V635

Cas. Astron. Astrophys., 439:1131–1134.

Becker, P. A. (1998). Dynamical Structure of Radiation-dominated Pulsar Accretion

Shocks. Astrophys. J., 498:790–801.

Becker, P. A. (2003). Exact solution for the Green’s function describing time-

dependent thermal Comptonization. Mon. Not. Roy. Astron. Soc., 343:215–240.

Becker, P. A., Klochkov, D., Schönherr, G., Nishimura, O., Ferrigno, C., Caballero,

I., Kretschmar, P., Wolff, M. T., Wilms, J., and Staubert, R. (2012). Spectral

formation in accreting X-ray pulsars: bimodal variation of the cyclotron energy

with luminosity. Astron. Astrophys., 544:A123.

114 References

Becker, P. A. and Wolff, M. T. (2007). Thermal and Bulk Comptonization in

Accretion-powered X-Ray Pulsars. Astrophys. J., 654:435–457.

Bhalerao, V., Romano, P., Tomsick, J., Natalucci, L., Smith, D. M., Bellm, E., Boggs,

S. E., Chakrabarty, D., Christensen, F. E., Craig, W. W., Fuerst, F., Hailey, C. J.,

Harrison, F. A., Krivonos, R. A., Lu, T.-N., Madsen, K., Stern, D., Younes, G., and

Zhang, W. (2015). NuSTAR detection of a cyclotron line in the supergiant fast

X-ray transient IGR J17544-2619. Mon. Not. Roy. Astron. Soc., 447:2274–2281.

Boldin, P. A., Tsygankov, S. S., and Lutovinov, A. A. (2013). On timing and spectral

characteristics of the X-ray pulsar 4U 0115+63: Evolution of the pulsation period

and the cyclotron line energy. Astronomy Letters, 39:375–388.

Borozdin, K., Gilfanov, M., Sunyaev, R., Churazov, E., Loznikov, V., Yamburenko,

N., Skinner, G. K., Patterson, T. G., Willmore, A. P., Emam, O., Brinkman, A. C.,

Heise, J., Int-Zand, J. J. M., and Jager, R. (1990). KS:1947+300 - a New Transient

X-Ray Source in Cygnus. Soviet Astronomy Letters, 16:345.

Burnard, D. J., Arons, J., and Klein, R. I. (1991). Accretion powered pulsars -

Continuum spectra and light curves of settling accretion mounds. Astrophys. J.,

367:575–592.

Caballero, I., Kraus, U., Santangelo, A., Sasaki, M., and Kretschmar, P. (2011).

Analyzing X-ray pulsar profiles: geometry and beam pattern of A 0535+26.

Astron. Astrophys., 526:A131.

Caballero, I. and Wilms, J. (2012). X-ray pulsars: a review. Memorie della Societa

Astronomica Italiana, 83:230.

Chakrabarty, D., Koh, T., Bildsten, L., Prince, T. A., Finger, M. H., Wilson, R. B.,

Pendleton, G. N., and Rubin, B. C. (1995). Discovery of the 18.7 Second Accreting

X-Ray Pulsar GRO J1948+32. Astrophys. J., 446:826.

Chakrabarty, D., Prince, T. A., Finger, M. H., Wilson, R. B., and Pendleton, G. N.

(1994). GRO J1948+32 = GRO J2014+34. IAU Circ., 5981.

References 115

Chandrasekhar, S. (1931). The Maximum Mass of Ideal White Dwarfs. Astrophys.

J., 74:81.

Coe, M. J., Payne, B. J., Longmore, A., and Hanson, C. G. (1988). The optical/IR

counterpart to the newly-discovered X-ray source EXO 2030+375. Mon. Not.

Roy. Astron. Soc., 232:865–871.

Corbet, R. H. D. (1986). The three types of high-mass X-ray pulsator. Mon. Not.

Roy. Astron. Soc., 220:1047–1056.

D’Aì, A., Cusumano, G., Del Santo, M., La Parola, V., and Segreto, A. (2017). A

broad-band self-consistent modelling of the X-ray spectrum of 4U 1626-67. Mon.

Not. Roy. Astron. Soc., 470:2457–2468.

Davidson, K. and Ostriker, J. P. (1973). Neutron-Star Accretion in a Stellar Wind:

Model for a Pulsed X-Ray Source. Astrophys. J., 179:585–598.

DeCesar, M. E., Boyd, P. T., Pottschmidt, K., Wilms, J., Suchy, S., and Miller, M. C.

(2013). The Be/X-Ray Binary Swift J1626.6-5156 as a Variable Cyclotron Line

Source. Astrophys. J., 762:61.

Epili, P., Naik, S., and Jaisawal, G. K. (2016). Broad-band spectroscopy of the tran-

sient X-ray binary pulsar KS 1947+300 during 2013 giant outburst: Detection of

pulsating soft X-ray excess component. Research in Astronomy and Astrophysics,

16:77.

Epili, P., Naik, S., Jaisawal, G. K., and Gupta, S. (2017). Decade long RXTE

monitoring observations of Be/X-ray binary pulsar EXO 2030+375. Mon. Not.

Roy. Astron. Soc., 472:3455–3466.

Ferrigno, C., Becker, P. A., Segreto, A., Mineo, T., and Santangelo, A. (2009). Study

of the accreting pulsar 4U 0115+63 using a bulk and thermal Comptonization

model. Astron. Astrophys., 498:825–836.

Ferrigno, C., Falanga, M., Bozzo, E., Becker, P. A., Klochkov, D., and Santangelo,

A. (2011). 4U 0115+63: phase lags and cyclotron resonant scattering. Astron.

Astrophys., 532:A76.

116 References

Ferrigno, C., Pjanka, P., Bozzo, E., Klochkov, D., Ducci, L., and Zdziarski, A. A.

(2016). Glancing through the accretion column of EXO 2030+375. Astron.

Astrophys., 593:A105.

Finger, M. H. and Prince, T. A. (1997). Recent results from observations of accreting

pulsars. In Dermer, C. D., Strickman, M. S., and Kurfess, J. D., editors, Pro-

ceedings of the Fourth Compton Symposium, volume 410 of American Institute of

Physics Conference Series, pages 57–74.

Finger, M. H., Stollberg, M., Pendleton, G. N., Wilson, R. B., Chakrabarty, D., Chiu,

J., and Prince, T. A. (1994). GRO J2014+34. IAU Circ., 5977.

Finger, M. H., Wilson, R. B., and Harmon, B. A. (1996). Quasi-periodic Oscillations

during a Giant Outburst of A0535+262. Astrophys. J., 459:288.

Frank, J., King, A., and Raine, D. J. (2002). Accretion Power in Astrophysics: Third

Edition.

Fraser, G. W. (1989). X-ray detectors in astronomy.

Fuerst, F., Wilson-Hodge, C. A., Kretschmar, P., Kajava, J., and Kuehnel, M. (2016).

Fading outbursts of EXO 2030+375. The Astronomer’s Telegram, 8835.

Fürst, F., Pottschmidt, K., Wilms, J., Kennea, J., Bachetti, M., Bellm, E., Boggs,

S. E., Chakrabarty, D., Christensen, F. E., Craig, W. W., Hailey, C. J., Harrison,

F., Stern, D., Tomsick, J. A., Walton, D. J., and Zhang, W. (2014). NuSTAR

Discovery of a Cyclotron Line in KS 1947+300. Astrophys. J. Lett., 784:L40.

Galloway, D. K., Morgan, E. H., and Levine, A. M. (2004). A Frequency Glitch in

an Accreting Pulsar. Astrophys. J., 613:1164–1172.

Ghosh, P. and Lamb, F. K. (1979). Accretion by rotating magnetic neutron stars. III -

Accretion torques and period changes in pulsating X-ray sources. Astrophys. J.,

234:296–316.

Giacconi, R., Gursky, H., Paolini, F. R., and Rossi, B. B. (1962). Evidence for x Rays

From Sources Outside the Solar System. Physical Review Letters, 9:439–443.

References 117

Giacconi, R., Gursky, H., and van Speybroeck, L. P. (1968). Observational Tech-

niques in X-Ray Astronomy. ARA&A, 6:373.

Gold, T. (1968). Rotating Neutron Stars as the Origin of the Pulsating Radio Sources.

Nature, 218:731–732.

Haberl, F., Angelini, L., Motch, C., and White, N. E. (1998). X-ray observations

of the slowest known Be/X-ray pulsars RXJ0146.9+6121 and XPersei. Astron.

Astrophys., 330:189–194.

Harding, A. K. (2013). The neutron star zoo. Frontiers of Physics, 8:679–692.

Harrison, F. A., Craig, W. W., Christensen, F. E., Hailey, C. J., Zhang, W. W., Boggs,

S. E., Stern, D., Cook, W. R., Forster, K., Giommi, P., Grefenstette, B. W., Kim,

Y., Kitaguchi, T., Koglin, J. E., Madsen, K. K., Mao, P. H., Miyasaka, H., Mori, K.,

Perri, M., Pivovaroff, M. J., Puccetti, S., Rana, V. R., Westergaard, N. J., Willis, J.,

Zoglauer, A., An, H., Bachetti, M., Barrière, N. M., Bellm, E. C., Bhalerao, V.,

Brejnholt, N. F., Fuerst, F., Liebe, C. C., Markwardt, C. B., Nynka, M., Vogel, J. K.,

Walton, D. J., Wik, D. R., Alexander, D. M., Cominsky, L. R., Hornschemeier,

A. E., Hornstrup, A., Kaspi, V. M., Madejski, G. M., Matt, G., Molendi, S., Smith,

D. M., Tomsick, J. A., Ajello, M., Ballantyne, D. R., Balokovic, M., Barret,

D., Bauer, F. E., Blandford, R. D., Brandt, W. N., Brenneman, L. W., Chiang,

J., Chakrabarty, D., Chenevez, J., Comastri, A., Dufour, F., Elvis, M., Fabian,

A. C., Farrah, D., Fryer, C. L., Gotthelf, E. V., Grindlay, J. E., Helfand, D. J.,

Krivonos, R., Meier, D. L., Miller, J. M., Natalucci, L., Ogle, P., Ofek, E. O.,

Ptak, A., Reynolds, S. P., Rigby, J. R., Tagliaferri, G., Thorsett, S. E., Treister, E.,

and Urry, C. M. (2013). The Nuclear Spectroscopic Telescope Array (NuSTAR)

High-energy X-Ray Mission. Astrophys. J., 770:103.

Hewish, A., Bell, S. J., Pilkington, J. D. H., Scott, P. F., and Collins, R. A. (1968).

Observation of a Rapidly Pulsating Radio Source. Nature, 217:709–713.

Hickox, R. C., Narayan, R., and Kallman, T. R. (2004). Origin of the Soft Excess in

X-Ray Pulsars. Astrophys. J., 614:881–896.

118 References

Jahoda, K., Swank, J. H., Giles, A. B., Stark, M. J., Strohmayer, T., Zhang, W., and

Morgan, E. H. (1996). In-orbit performance and calibration of the Rossi X-ray

Timing Explorer (RXTE) Proportional Counter Array (PCA). In Siegmund, O. H.

and Gummin, M. A., editors, EUV, X-Ray, and Gamma-Ray Instrumentation for

Astronomy VII, volume 2808 of , pages 59–70.

Jaisawal, G. K. and Naik, S. (2017). Cyclotron absorption lines in the era of Suzaku

and NuSTAR. In Serino, M., Shidatsu, M., Iwakiri, W., and Mihara, T., editors, 7

years of MAXI: monitoring X-ray Transients, held 5-7 December 2016 at RIKEN.

Online at https://indico2.riken.jp/indico/conferenceDisplay.py?confId=2357,

p.153, page 153.

Jaisawal, G. K., Naik, S., and Epili, P. (2016). Suzaku view of the Be/X-ray binary

pulsar GX 304-1 during Type I X-ray outbursts. Mon. Not. Roy. Astron. Soc.,

457:2749–2760.

Jaisawal, G. K., Naik, S., and Paul, B. (2013). Possible Detection of a Cyclotron

Resonance Scattering Feature in the X-Ray Pulsar 4U 1909+07. Astrophys. J.,

779:54.

James, M., Paul, B., Devasia, J., and Indulekha, K. (2010). Discovery of a 0.02 Hz

QPO feature in the transient X-ray pulsar KS 1947+300. Mon. Not. Roy. Astron.

Soc., 407:285–290.

Kalogera, V. and Baym, G. (1996). The Maximum Mass of a Neutron Star. Astrophys.

J. Lett., 470:L61.

Kawagoe, A., Mihara, T., Negoro, H., Nakajima, M., Nakahira, S., Ueno, S., Tomida,

H., Kimura, M., Ishikawa, M., Nakagawa, Y. E., Sugizaki, M., Morii, M., Serino,

M., Sugimoto, J., Takagi, T., Matsuoka, M., Kawai, N., Usui, R., Ishikawa, K.,

Yoshii, T., Yoshida, A., Sakamoto, T., Nakano, Y., Tsunemi, H., Sasaki, M.,

Sakakibara, H., Fukushima, K., Onodera, T., Suzuki, K., Ueda, Y., Shidatsu, M.,

Kawamuro, T., Tsuboi, Y., Higa, M., Yamauchi, M., Yoshidome, K., Ogawa,

Y., Yamada, H., and Yamaoka, K. (2013). MAXI/GSC detection of a renewed

References 119

X-ray activity of the Be transient X-ray pulsar KS 1947+300. The Astronomer’s

Telegram, 5438.

Kennea, J. A., Evans, P. A., Krimm, H. A., Romano, P., Mangano, V., Curran, P.,

Yamaoka, K., and Negoro, H. (2013). Swift confirms KS 1947+300 is in X-ray

outburst. The Astronomer’s Telegram, 5441.

Klochkov, D., Santangelo, A., Staubert, R., and Ferrigno, C. (2008). Giant outburst

of EXO 2030+375: pulse-phase resolved analysis of INTEGRAL data. Astron.

Astrophys., 491:833–840.

Koyama, K., Tsunemi, H., Dotani, T., Bautz, M. W., Hayashida, K., Tsuru, T. G.,

Matsumoto, H., Ogawara, Y., Ricker, G. R., Doty, J., Kissel, S. E., Foster, R.,

Nakajima, H., Yamaguchi, H., Mori, H., Sakano, M., Hamaguchi, K., Nishiuchi,

M., Miyata, E., Torii, K., Namiki, M., Katsuda, S., Matsuura, D., Miyauchi,

T., Anabuki, N., Tawa, N., Ozaki, M., Murakami, H., Maeda, Y., Ichikawa, Y.,

Prigozhin, G. Y., Boughan, E. A., Lamarr, B., Miller, E. D., Burke, B. E., Gregory,

J. A., Pillsbury, A., Bamba, A., Hiraga, J. S., Senda, A., Katayama, H., Kitamoto,

S., Tsujimoto, M., Kohmura, T., Tsuboi, Y., and Awaki, H. (2007). X-Ray Imaging

Spectrometer (XIS) on Board Suzaku. Pub. Astron. Soc. Japan, 59:23–33.

Kraus, U., Nollert, H.-P., Ruder, H., and Riffert, H. (1995). Analyzing X-Ray Pulsar

Profiles: Asymmetry as a Key to Geometry and Beam Pattern. Astrophys. J.,

450:763.

Kretschmar, P. (1996). Hochenergieröntgenspektren der akkretierenden Röntgenpul-

sare Vela X-1 und A 0535+26. Zenodo. In German.

Kretschmar, P., Fuerst, F., Wilson-Hodge, C. A., Jenke, P., Kajava, J., Kuehnel,

M., Wilms, J., Blay, P., Pottschmidt, K., and Alfonso-Garzon, J. (2016). EXO

2030+375 restarts in reverse. The Astronomer’s Telegram, 9485.

Krimm, H., Barthelmy, S., Gehrels, N., Markwardt, C., Palmer, D., Sanwal, D.,

and Tueller, J. (2006). Swift/BAT detects an outburst from EXO 2030+375. The

Astronomer’s Telegram, 861.

120 References

Lattimer, J. M. (2012). The Nuclear Equation of State and Neutron Star Masses.

Annual Review of Nuclear and Particle Science, 62:485–515.

Lattimer, J. M. and Prakash, M. (2004). The physics of neutron stars. Science,

304(5670):536–542.

Leahy, D. A., Darbro, W., Elsner, R. F., Weisskopf, M. C., Sutherland, P. G., Kahn, S.,

and Grindlay, J. E. (1983a). On searches for pulsed emission with application to

four globular cluster X-ray sources - NGC 1851, 6441, 6624, and 6712. Astrophys.

J., 266:160–170.

Leahy, D. A., Elsner, R. F., and Weisskopf, M. C. (1983b). On searches for periodic

pulsed emission - The Rayleigh test compared to epoch folding. Astrophys. J.,

272:256–258.

Levine, A. and Corbet, R. (2000). KS 1947+300. IAU Circ., 7523.

Lewin, W. H. G. and van der Klis, M. (2006). Compact Stellar X-ray Sources.

Liu, Q. Z., van Paradijs, J., and van den Heuvel, E. P. J. (2006). Catalogue of high-

mass X-ray binaries in the Galaxy (4th edition). Astron. Astrophys., 455:1165–

1168.

Liu, Q. Z., van Paradijs, J., and van den Heuvel, E. P. J. (2007). A catalogue of

low-mass X-ray binaries in the Galaxy, LMC, and SMC (Fourth edition). Astron.

Astrophys., 469:807–810.

Manchester, R. N., Hobbs, G. B., Teoh, A., and Hobbs, M. (2005). The Australia

Telescope National Facility Pulsar Catalogue. Astron. J., 129:1993–2006.

Mihalas, D. and Winkler, K.-H. A., editors (1986). in Radiation Hydrodynamics

in Stars and Compact Objects, volume 255 of Lecture Notes in Physics, Berlin

Springer Verlag.

Mihara, T. (1995). Observational study of X-ray spectra of binary pulsars with

Ginga. PhD thesis, , Dept. of Physics, Univ. of Tokyo (M95), (1995).

References 121

Miller, M. C. and Miller, J. M. (2015). The masses and spins of neutron stars and

stellar-mass black holes. Physics Reports, 548:1–34.

Motch, C. and Janot-Pacheco, E. (1987). The optical counterpart of the X-ray

transient EXO 2030+375. Astron. Astrophys., 182:L55–L58.

Mushtukov, A. A., Suleimanov, V. F., Tsygankov, S. S., and Poutanen, J. (2015).

The critical accretion luminosity for magnetized neutron stars. Mon. Not. Roy.

Astron. Soc., 447:1847–1856.

Nagase, F. (1989). Accretion-powered X-ray pulsars. Pub. Astron. Soc. Japan,

41:1–79.

Naik, S. (2013). X-ray and infrared properties of Be/X-ray binary pulsars. In

Das, S., Nandi, A., and Chattopadhyay, I., editors, Astronomical Society of India

Conference Series, volume 8 of Astronomical Society of India Conference Series,

pages 103–108.

Naik, S., Callanan, P. J., Paul, B., and Dotani, T. (2006). Broadband X-Ray Spectrum

of KS 1947+300 with BeppoSAX. Astrophys. J., 647:1293–1299.

Naik, S., Dotani, T., Terada, Y., Nakajima, M., Mihara, T., Suzuki, M., Makishima,

K., Sudoh, K., Kitamoto, S., Nagase, F., Enoto, T., and Takahashi, H. (2008).

Broadband X-Ray Spectroscopy of A0535+262 with Suzaku. Astrophys. J.,

672:516–523.

Naik, S. and Jaisawal, G. K. (2015). Suzaku observation of Be/X-ray binary pulsar

EXO 2030+375. Research in Astronomy and Astrophysics, 15:537.

Naik, S. and Paul, B. (2002). Spectral Properties of the X-ray Binary Pulsar LMC

X-4 during Different Intensity States. Journal of Astrophysics and Astronomy,

23:27.

Naik, S. and Paul, B. (2004). Timing and Spectral Studies of LMC X-4 in High

and Low States with BeppoSAX: Detection of Pulsations in the Soft Spectral

Component. Astrophys. J., 600:351–357.

122 References

Naik, S., Paul, B., and Ali, Z. (2011a). X-Ray Spectroscopy of the High-mass X-Ray

Binary Pulsar Centaurus X-3 over Its Binary Orbit. Astrophys. J., 737:79.

Naik, S., Paul, B., Kachhara, C., and Vadawale, S. V. (2011b). Suzaku observation of

the transient X-ray pulsar GRO J1008-57. Mon. Not. Roy. Astron. Soc., 413:241–

248.

Negueruela, I. (1998). On the nature of Be/X-ray binaries. Astron. Astrophys.,

338:505–510.

Negueruela, I. (2007). Be/X-ray Binaries: An Observational Approach. In St.-Louis,

N. and Moffat, A. F. J., editors, Massive Stars in Interactive Binaries, volume 367

of Astronomical Society of the Pacific Conference Series, page 477.

Parmar, A. N., White, N. E., and Stella, L. (1989a). The transient 42 second X-

ray pulsar EXO 2030+375. II - The luminosity dependence of the pulse profile.

Astrophys. J., 338:373–380.

Parmar, A. N., White, N. E., Stella, L., Izzo, C., and Ferri, P. (1989b). The transient

42 second X-ray pulsar EXO 2030+375. I - The discovery and the luminosity

dependence of the pulse period variations. Astrophys. J., 338:359–372.

Paul, B. and Naik, S. (2011). Transient High Mass X-ray Binaries. Bulletin of the

Astronomical Society of India, 39:429–449.

Porter, J. M. and Rivinius, T. (2003). Classical Be Stars. Pub. Astron. Soc. Pac.,

115:1153–1170.

Postnov, K. A., Mironov, A. I., Lutovinov, A. A., Shakura, N. I., Kochetkova, A. Y.,

and Tsygankov, S. S. (2015). Spin-up/spin-down of neutron star in Be-X-ray

binary system GX 304-1. Mon. Not. Roy. Astron. Soc., 446:1013–1019.

Pottschmidt, K., Kreykenbohm, I., Wilms, J., Coburn, W., Rothschild, R. E.,

Kretschmar, P., McBride, V., Suchy, S., and Staubert, R. (2005). RXTE Discovery

of Multiple Cyclotron Lines during the 2004 December Outburst of V0332+53.

Astrophys. J. Lett., 634:L97–L100.

References 123

Ramsey, B. D., Alexander, C. D., Apple, J. A., Benson, C. M., Dietz, K. L., Elsner,

R. F., Engelhaupt, D. E., Ghosh, K. K., Kolodziejczak, J. J., O’Dell, S. L., Speegle,

C. O., Swartz, D. A., and Weisskopf, M. C. (2002). First Images from HERO, a

Hard X-Ray Focusing Telescope. Astrophys. J., 568:432–435.

Reig, P. (2011). Be/X-ray binaries. Astrophys. Space Sci., 332:1–29.

Reig, P. and Roche, P. (1999). Discovery of two new persistent Be/X-ray pulsar

systems. Mon. Not. Roy. Astron. Soc., 306:100–106.

Reynolds, A. P., Parmar, A. N., and White, N. E. (1993). The luminosity depen-

dence of the X-ray spectrum of the transient 42 second pulsar EXO 2030 + 375.

Astrophys. J., 414:302–309.

Rhoades, C. E. and Ruffini, R. (1974). Maximum Mass of a Neutron Star. Physical

Review Letters, 32:324–327.

Sasaki, M., Klochkov, D., Kraus, U., Caballero, I., and Santangelo, A. (2010).

Analyzing X-ray pulsar profiles: geometry and beam pattern of EXO 2030+375.

Astron. Astrophys., 517:A8.

Sasaki, M., Müller, D., Kraus, U., Ferrigno, C., and Santangelo, A. (2012). Analysing

X-ray pulsar profiles. Geometry and beam pattern of 4U 0115+63 and V 0332+53.

Astron. Astrophys., 540:A35.

Schönherr, G., Schwarm, F.-W., Falkner, S., Dauser, T., Ferrigno, C., Kühnel,

M., Klochkov, D., Kretschmar, P., Becker, P. A., Wolff, M. T., Pottschmidt, K.,

Falanga, M., Kreykenbohm, I., Fürst, F., Staubert, R., and Wilms, J. (2014). For-

mation of phase lags at the cyclotron energies in the pulse profiles of magnetized,

accreting neutron stars. Astron. Astrophys., 564:L8.

Schönherr, G., Wilms, J., Kretschmar, P., Kreykenbohm, I., Santangelo, A., Roth-

schild, R. E., Coburn, W., and Staubert, R. (2007). A model for cyclotron reso-

nance scattering features. Astron. Astrophys., 472:353–365.

124 References

Shakura, N., Postnov, K., Kochetkova, A., and Hjalmarsdotter, L. (2012). Theory of

quasi-spherical accretion in X-ray pulsars. Mon. Not. Roy. Astron. Soc., 420:216–

236.

Stollberg, M. T. (1997). Determining the Orbit of the X-Ray Transient EXO

2030+375 with the Burst and Transient Source Experiment. PhD thesis, UNIVER-

SITY OF ALABAMA IN HUNTSVILLE.

Suchy, S., Pottschmidt, K., Wilms, J., Kreykenbohm, I., Schönherr, G., Kretschmar,

P., McBride, V., Caballero, I., Rothschild, R. E., and Grinberg, V. (2008). Pulse

Phase-Resolved Analysis of the High-Mass X-Ray Binary Centaurus X-3 over

Two Binary Orbits. Astrophys. J., 675:1487–1498.

Sun, X.-J., Li, T.-P., Wu, M., and Cheng, L.-X. (1994). The X-ray spectrum of EXO

2030 + 375. Astron. Astrophys., 289:127–130.

Sunyaev, R. A. and Titarchuk, L. G. (1980). Comptonization of X-rays in plasma

clouds - Typical radiation spectra. Astron. Astrophys., 86:121–138.

Swank, J. and Morgan, E. (2000). KS 1947+300 = GRO J1948+32. IAU Circ., 7531.

Taam, R. E. and van den Heuvel, E. P. J. (1986). Magnetic field decay and the origin

of neutron star binaries. Astrophys. J., 305:235–245.

Tauris, T. M. and van den Heuvel, E. P. J. (2006). Formation and evolution of

compact stellar X-ray sources, pages 623–665.

Torrejón, J. M., Schulz, N. S., Nowak, M. A., and Kallman, T. R. (2010). A Chandra

Survey of Fluorescence Fe Lines in X-ray Binaries at High Resolution. Astrophys.

J., 715:947–958.

Truemper, J. (1978). Cyclotron lines from HER X-1. Technical report.

Tsygankov, S. S. and Lutovinov, A. A. (2005). Observations of the Transient X-ray

Pulsar KS 1947+300 by the INTEGRAL and RXTE Observatories. Astronomy

Letters, 31:88–97.

References 125

Tsygankov, S. S., Lutovinov, A. A., Doroshenko, V., Mushtukov, A. A., Suleimanov,

V., and Poutanen, J. (2016). Propeller effect in two brightest transient X-ray

pulsars: 4U 0115+63 and V 0332+53. Astron. Astrophys., 593:A16.

Walter, R., Lutovinov, A. A., Bozzo, E., and Tsygankov, S. S. (2015). High-mass X-

ray binaries in the Milky Way. A closer look with INTEGRAL. Astron. Astrophys.

Rev., 23:2.

Waters, L. B. F. M., van den Heuvel, E. P. J., Taylor, A. R., Habets, G. M. H. J., and

Persi, P. (1988). Evidence for low-velocity winds in Be/X-ray binaries. Astron.

Astrophys., 198:200–210.

White, N. E., Swank, J. H., and Holt, S. S. (1983). Accretion powered X-ray pulsars.

Astrophys. J., 270:711–734.

Wilson, C. A., Fabregat, J., and Coburn, W. (2005). Discovery of a Transition to

Global Spin-up in EXO 2030+375. Astrophys. J. Lett., 620:L99–L102.

Wilson, C. A., Finger, M. H., and Camero-Arranz, A. (2008). Outbursts Large and

Small from EXO 2030+375. Astrophys. J., 678:1263–1272.

Wilson, C. A., Finger, M. H., Coe, M. J., Laycock, S., and Fabregat, J. (2002). A

Decade in the Life of EXO 2030+375: A Multiwavelength Study of an Accreting

X-Ray Pulsar. Astrophys. J., 570:287–302.

Wolff, M. T., Becker, P. A., Gottlieb, A. M., Fürst, F., Hemphill, P. B., Marcu-

Cheatham, D. M., Pottschmidt, K., Schwarm, F.-W., Wilms, J., and Wood, K. S.

(2016). The NuSTAR X-Ray Spectrum of Hercules X-1: A Radiation-dominated

Radiative Shock. Astrophys. J., 831:194.

Wolter, H. (1952a). Spiegelsysteme streifenden einfalls als abbildende optiken für

röntgenstrahlen. Annalen der Physik, 445(12):94–114.

Wolter, H. (1952b). Verallgemeinerte schwarzschildsche spiegelsysteme streifender

reflexion als optiken für röntgenstrahlen. Annalen der Physik, 445(45):286–295.

List of Publications

A Refereed Journal Publications included in the Thesis

1. “Broad-band spectroscopy of the transient X-ray binary pulsar KS 1947+300

during 2013 giant outburst: Detection of pulsating soft X-ray excess com-

ponent”

Epili, P., Naik, S., Jaisawal, G. K.

2016, Research in Astronomy and Astrophysics, 16, 5

2. “Decade long RXTE monitoring observations of Be/X-ray binary pulsar

EXO 2030+375”

Epili, P., Naik, S., Jaisawal, G. K.

2017, Monthly Notices of Royal Astronomical Society, 472, 3455

B Additional Publications

1. “Suzaku view of Be/X-ray binary pulsar GX 304-1 during Type-I X-ray

outbursts”

Jaisawal, G. K., Naik, S., Epili, P.,

2016, Monthly Notices of Royal Astronomical Society, 457, 2749

2. “A curious case of the accretion-powered X-ray pulsar GX 1+4”

Jaisawal, G. K., Naik, S., Gupta, S., Jérôme Chenevez, Epili, P.

2018, Monthly Notices of Royal Astronomical Society, (in press) .

128 References

C Presentations at Conferences and Symposia

1. Contributed Oral presentation on ‘Decade long RXTE monitoring observations

of Be/X-ray binary pulsar EXO 2030+375’ at the workshop Be/X-ray Binaries

- 2017 held at Heraklion (Greece) during September 11-13, 2017.

2. Presented a poster on ‘A Comprehensive timing and spectral studies of Be/X-

ray binary pulsar EXO 2030+375 in the RXTE era’ at a AHEAD project

meeting on “The Power of X-ray Spectroscopy” held at Staszic Palace, Polish

Academy of Sciences, Warsaw, Poland during September 06-08, 2017.

3. Contributed Oral presentation on ‘Timing and spectral studies of the transient

X-ray binary pulsar KS 1947+300’ at a conference on Wide Band Spectral

and Timing Studies of Cosmic X-ray Sources held at TIFR, Mumbai during

January 10-13, 2017.

4. Contributed Oral presentation on ‘Suzaku observations of KS 1947+300 during

a giant outburst in 2013’ at National Space Science Symposium 2016 held at

VSSC, Thiruvananthapuram, Kerala during February 9-12, 2016.

5. Presented a poster on ‘Suzaku observations of KS 1947+300 during an X-ray

outburst’ at ASI 2015 Meeting held at NCRA, Pune during February 17-20,

2015.

Publications attached with Thesis

1. “Decade long RXTE monitoring observations of Be/X-ray binary pulsar

EXO 2030+375”

Epili, P., Naik, S., Jaisawal, G. K.

2017, Monthly Notices of Royal Astronomical Society, 472, 3455

2. “Broad-band spectroscopy of the transient X-ray binary pulsar KS 1947+300

during 2013 giant outburst: Detection of pulsating soft X-ray excess com-

ponent”

Epili, P., Naik, S., Jaisawal, G. K.

2016, Research in Astronomy and Astrophysics, 16, 5

RAA 2016 Vol. 16 No. 5, 77 (10pp) doi: 10.1088/1674–4527/16/5/077

http://www.raa-journal.org http://iopscience.iop.org/raa

Research in

Astronomy andAstrophysics

Broad-band spectroscopy of the transient X-ray binary pulsar KS 1947+300

during 2013 giant outburst: Detection of pulsating soft X-ray excess component

Prahlad Epili, Sachindra Naik and Gaurava K. Jaisawal

Astronomy and Astrophysics Division, Physical Research Laboratory, Ahmedabad 380009, India; prahlad@prl.res.in

(PE); snaik@prl.res.in (SN); gaurava@prl.res.in (GKJ)

Received 2015 August 19; accepted 2016 February 11

Abstract We present the results obtained from detailed timing and spectral studies of the Be/X-ray binary

pulsar KS 1947+300 during its 2013 giant outburst. We used data from Suzaku observations of the pulsar

at two epochs, i.e. on 2013 October 22 (close to the peak of the outburst) and 2013 November 22. X-

ray pulsations at ∼18.81 s were clearly detected in the light curves obtained from both observations. Pulse

periods estimated during the outburst showed that the pulsar was spinning up. The pulse profile was found to

be single-peaked up to ∼10 keV beyond which a sharp peak followed by a dip-like feature appeared at hard

X-rays. The dip-like feature has been observed up to ∼70 keV. The 1–110 keV broad-band spectroscopy

of both observations revealed that the best-fit model was comprised of a partially absorbed Negative and

Positive power law with EXponential cutoff (NPEX) continuum model along with a blackbody component

for the soft X-ray excess and two Gaussian functions at 6.4 and 6.7 keV for emission lines. Both the

lines were identified as emission from neutral and He-like iron atoms. To fit the spectra, we included the

previously reported cyclotron absorption line at 12.2 keV. From the spin-up rate, the magnetic field of the

pulsar was estimated to be ∼1.2×1012 G and found to be comparable to that obtained from the detection

of the cyclotron absorption feature. Pulse-phase resolved spectroscopy revealed the pulsating nature of the

soft X-ray excess component in phase with the continuum flux. This confirms that the accretion column

and/or accretion stream are the most probable regions of the soft X-ray excess emission in KS1947+300.

The presence of the pulsating soft X-ray excess in phase with continuum emission may be the possible

reason for not observing the dip at soft X-rays.

Key words: pulsars: individual (KS 1947+300) — stars: neutron — X-rays: binaries

1 INTRODUCTION

Be/X-ray binaries (BeXBs) are known to be the largest

subclass (∼60%) of high-mass X-ray binaries (Caballero

& Wilms 2012). The majority of BeXBs consist of a neu-

tron star as a compact object and a Be star as an optical

companion. The optical companions in these binary sys-

tems are non-supergiant B-type stars of luminosity class

III-V that show emission lines in their optical/infrared

spectra (Okazaki & Negueruela 2001; Reig 2011). The

neutron star in BeXBs accretes matter from the circum-

stellar disk of the Be star, usually at the periastron pas-

sage. This abrupt accretion of a huge amount of matter

causes significant enhancement of the X-ray emission from

the pulsating neutron star. These periodic enhancements of

the X-ray intensity are known as Type I X-ray outbursts

(peak luminosity ∼ 1035− 1037 erg s−1). The neutron star

in these systems, however, occasionally shows rare X-ray

outbursts (Type II), lasting for several tens of days to a few

months during which the peak luminosity reaches up to

1038 erg s−1. For a brief review of the properties of BeXBs,

refer to Paul & Naik (2011).

The transient BeXB pulsar KS 1947+300 was discov-

ered on 1989 June 8 with the Kvant/TTM coded-mask X-

ray spectrometer on the Mir space station (Skinner 1989;

Borozdin et al. 1990). The pulsar was first detected at a

flux level of 70 mCrab that later decreased to ∼10 mCrab

within 2 months of the detection. The spectra obtained

from these observations were described by an absorbed

power-law with a photon index of 1.72±0.31 (Borozdin

et al. 1990).

In April 1994, the X-ray pulsar GRO J1948+32 was

discovered close to the coordinates of KS 1947+300 by

the Burst and Transient Source Experiment (BATSE) in-

strument onboard the Compton Gamma Ray Observatory

(CGRO) (Chakrabarty et al. 1995). The pulsation from

this new source was found to be 18.7 s. Based on the

spin period analysis, later Swank & Morgan (2000) estab-

lished that KS 1947+300 and GRO J1948+32 are the same

source. A Be star with a visible magnitude of 14.2 at a dis-

77–2 P. Epili, S. Naik & G. K. Jaisawal

tance of ∼10 kpc was discovered as the optical counterpart

of the pulsar (Negueruela et al. 2003).

KS 1947+300 was inactive from 1995 to 2000 without

showing any major X-ray outburst. Subsequent RXTE ob-

servations showed that the source became active in October

2000 and went into a strong X-ray outburst (Levine &

Corbet 2000). The pulsar spectrum in the 2–80 keV range

during the outburst was described with a Comptonization

continuum model along with a blackbody component for

the soft X-ray excess and a Gaussian component for the

iron emission line at 6.5 keV (Galloway et al. 2004). The

orbital parameters of the binary system were also esti-

mated from observations during October and are reported

in Galloway et al. (2004). A glitch in the pulsar frequency,

generally seen in anomalous X-ray pulsars and radio pul-

sars, but which is rare in accretion powered X-ray pul-

sars, was first detected in KS 1947+300 (Galloway et al.

2004). A low frequency quasi-periodic oscillation (QPO)

at 20 mHz was detected on several occasions in the de-

clining phase of the 2001 outburst with RXTE (James et al.

2010). Detection of low frequency QPOs and strong pulsa-

tions at low luminosity levels indirectly indicated that the

magnetic field of the neutron star is < 1013 G as predicated

from spin-up torque and luminosity correlation (James

et al. 2010). Using BeppoSAX observations of the 2001 X-

ray outburst, Naik et al. (2006) described the broad-band

pulsar spectrum in 0.1–100 keV with a Comptonization

model along with a blackbody component at ∼0.6 keV

and detected a weak 6.7 keV emission line from helium-

like iron atoms. However, the iron Kα emission line was

absent in the spectrum during the 2001 outburst.

A series of weak outbursts was observed from

KS 1947+300 during the period of 2002–2004. Among

these outbursts, the strongest one was detected with

INTEGRAL in April 2004. A high energy cutoff power-

law model was used to describe the spectra obtained from

INTEGRAL/ISGRI and JEM-X data during this outburst

(Tsygankov & Lutovinov 2005). However, there was no

signature of cyclotron absorption lines in the pulsar spec-

tra. In October 2013, KS 1947+300 was detected in a giant

outburst with a peak flux of 130 mCrab in the 3–10 keV

range (Furst et al. 2014). Several Swift/XRT and three

NuSTAR pointed observations were performed at differ-

ent phases of the X-ray outburst. Combined spectra from

Swift/XRT and NuSTAR in the 0.8–79 keV range were de-

scribed by a power law with an exponential cutoff con-

tinuum model along with a blackbody and an iron line

component at 6.5 keV (Furst et al. 2014). A cyclotron ab-

sorption line at ∼12.2 keV was discovered in the pulsar

spectra only during the second NuSTAR observation. The

surface magnetic field of the pulsar was estimated to be

∼1.1×1012(1 + z) G. Although KS 1947+300 had gone

through several major X-ray outbursts, the cyclotron ab-

sorption line was not detected in the spectra obtained from

earlier RXTE, BeppoSAX or INTEGRAL observations.

As the October 2013 outburst was a giant outburst, the

pulsar was active for a few months during which it was

5.655×104 5.66×104 5.665×104 5.67×104

0

0.05

Co

unts

/cm

2/s

ec

(15−

50

keV

)

Time (MJD)

Observation 1

Observation 2

(2013 October 22)

(2013 November 22)

Fig. 1 Swift/BAT light curve of KS 1947+300 in the 15–50 keV

energy range from 2013 August 14 (MJD 56518) to 2014 March

03 (MJD 56719). The arrow marks in the figure show the dates of

Suzaku observations of the pulsar during the giant X-ray outburst.

observed with different X-ray observatories. We have used

two Suzaku observations of the pulsar during this outburst

to study its broad-band timing and spectral properties. The

details on the observations, analysis, results and conclu-

sions are presented in the following sections of the paper.

2 OBSERVATIONS AND DATA ANALYSIS

The fifth Japanese X-ray satellite, Suzaku, was launched

on 2005 July 10 (Mitsuda et al. 2007). The X-ray imag-

ing spectrometers (XISs; Koyama et al. 2007) and the

hard X-ray detectors (HXDs; Takahashi et al. 2007) on-

board Suzaku covered the 0.5–600 keV range. The XISs

are CCD cameras located at the focal points of each of

the four X-ray telescopes (XRTs; Serlemitsos et al. 2007).

XIS-0, XIS-2 and XIS-3 are front-illuminated while XIS-1

is back-illuminated. The hard X-ray unit of Suzaku con-

sists of two non-imaging detectors, e.g. HXD/PIN and

HXD/GSO. HXD/PIN consists of silicon diode detectors

that work in the 10–70 keV range and HXD/GSO consists

of crystal scintillator detectors covering the 40–600 keV

range. The field of view of XIS is 17.8′′ × 17.8′′ in open

window mode. HXD/GSO has the same field of view of

34′′ × 34′′ as HXD/PIN up to 100 keV.

Target of opportunity observations of KS 1947+300

were carried out with Suzaku at two epochs during its giant

outburst in October-November 2013. The first observation

was performed on 2013 October 22, close to the peak of the

outburst for an effective exposure of ∼29 ks. The second

observation was carried out on 2013 November 22–23 at

the peak of the outburst for an effective exposure of ∼8 ks

and 32 ks for XIS and HXD, respectively. The XIS detec-

tors were operated in normal and burst clock mode with

2 s and 0.5 s time resolutions for the first and second ob-

servations, respectively. Both the observations were carried

out in the ‘XIS nominal’ position. A Swift/BAT monitoring

light curve of the pulsar in the 15–50 keV range covering

the giant X-ray outburst is shown in Figure 1. The arrow

marks in the figure represent the date of Suzaku observa-

Suzaku Observations of KS 1947+300 77–3

Table 1 Pulse Period History of KS 1947+300 during its 2013Giant Outburst

MJD Date Period (s) Reference

56586.79 2013–10–21 18.80584(16) Furst et al. (2014)

56587.22 2013–10–22 18.8088(1) present work

56618.61 2013–11–22 18.7878(1) present work

56618.91 2013–11–22 18.78399(7) Furst et al. (2014)

56635.75 2013–12–09 18.77088(6) Furst et al. (2014)

57053 2015–01–31 18.76255 Finger et al. (2015)

tions of the pulsar. In the present study, we used the pub-

licly available data with Observation IDs: 908001010 and

908001020 (hereafter Obs.1 and Obs.2 respectively).

2.1 Data Reduction

Unfiltered XIS and HXD event data were reprocessed by

using the aepipeline task in the Heasoft (version 6.16)

analysis package. Calibration database (CALDB) files re-

leased on 2012 November 06 (XIS) and 2011 September

13 (HXD) were used during data reprocessing. Cleaned

event files generated after reprocessing were used in the

present study. The aebarycen task of FTOOLS was ap-

plied on the cleaned event data to neutralize the effects

of motions of the satellite and the Earth around the Sun.

We checked the effect of thermal flexing by applying the

attitude correction S-lang script aeattcor.sl1 on data from

the XISs. After attitude correction, the XIS cleaned events

were examined for possible photon pile-up by using the

S-lang script pile estimate.sl2. During Obs.1, we detected

a pile-up of ∼31%, 21% and 33% at the centers of XIS-0,

XIS-1 and XIS-3 images, respectively. Therefore, an annu-

lus region with inner and outer radii of 75′′ and 200′′ was

chosen to reduce the pile-up below 4%. As in the case of

Obs.1, we estimated photon pile-up for Obs.2 which was

found to be ∼15%, 12% and 18% at the centers of XIS-0,

XIS-1 and XIS-3 images, respectively. An annulus region

with inner and outer radii of 35′′ and 200′′ was consid-

ered for the pile-up correction in the second observation.

The light-curves and spectra of the pulsar were extracted

from the XIS cleaned event data by applying the annu-

lus regions in the XSELECT package. Background light

curves and spectra were accumulated from a source free

region in the XIS image frame. Response and effective

area files for XISs were created from the “resp=yes” task

during the spectral extraction in XSELECT. HXD/PIN and

HXD/GSO source light curves and spectra were extracted

from the cleaned event data by applying the GTI selection

in XSELECT. The PIN and GSO background light curves

and spectra were generated in a similar manner from the

simulated tuned non-X-ray background event data pro-

vided by the instrument team. The response file released

in June 2011 was used for HXD/PIN for both the obser-

vations. However, GSO response and additional effective

1 http://space.mit.edu/ASC/software/suzaku/aeattcor.sl2 http://space.mit.edu/ASC/software/suzaku/pile estimate.sl

area files released on 2010 May 24 and 2010 May 26, re-

spectively, were used while analyzing HXD/GSO data.

3 TIMING ANALYSIS

Source and background light curves were extracted from

the reprocessed and barycentric corrected XISs, PIN and

GSO event data with time resolutions of 2 s, 1 s, 1 s for

Obs.1 and 0.5 s, 1 s, 1 s for Obs.2, respectively. The χ2-

maximization technique was applied to search for the

periodicity in the background subtracted XIS and PIN

light curves. Pulsations at periods of 18.8088(1) and

18.7878(1) s were detected in the light curves obtained

from the first and second Suzaku observations, respec-

tively. Though the observations were carried out within

a time interval of one month, the decrease of the pulse

period during a later epoch suggests that the pulsar was

spinning up. The pulse period history of KS 1947+300,

obtained from NuSTAR and Suzaku observations during

the October-November 2013 outburst, is given in Table 1.

Decreasing values of the spin period with time confirmed

that the pulsar was spinning-up during the outburst. A re-

cent measurement of pulse period of KS 1947+300 in the

January-February 2015 outburst (see Table 1) also indi-

cated the long term spin-up trend in the pulsar.

Pulse profiles of the pulsar in different energy bands

that were generated from the XIS, PIN and GSO light

curves, obtained from both observations, are shown in

Figures 2 and 3. During the first observation, the soft X-

ray pulse profile (below 5 keV) was found to be smooth

and single peaked. However, with the increase in energy,

a narrow dip-like feature appeared in the pulse profile and

became prominent in the 30–40 keV range. Beyond this en-

ergy, the depth of the dip decreased and disappeared from

the pulse profiles in the 70–100 keV range. During the sec-

ond observation, the pulse profiles were found to show a

similar type of energy dependence as seen during the first

observation. X-ray pulsations in the light curves were de-

tected up to ∼120 keV and ∼150 keV during the first and

second observations, respectively. Absorption dips in the

pulse profiles of BeXB pulsars during X-ray outbursts are

found to be prominent in soft X-rays but weak in hard X-

ray bands (Paul & Naik 2011 and references therein; Naik

et al. 2013; Naik & Jaisawal 2015). However, in the case

of Suzaku observations of KS 1947+300, the dip was found

to be absent in the soft X-ray pulse profiles and prominent

in the hard X-ray pulse profiles. It is, therefore, interesting

to investigate the spectral properties of the pulsar to under-

stand the causes of the absence/presence of an absorption

dip in soft/hard X-ray pulse profiles.

4 SPECTRAL ANALYSIS

4.1 Pulse-Phase Averaged Spectroscopy

We performed phase-averaged spectroscopy of

KS 1947+300 by using data from both Suzaku obser-

vations carried out during the giant outburst. Data from

77–4 P. Epili, S. Naik & G. K. Jaisawal

0.5

1

1.5

XIS−0

0.5−5 keV

0.5

1

XIS−0

5−7 keV

0.5

1

XIS−0

Norm

aliz

ed I

nte

nsi

ty

7−10 keV

0 0.5 1 1.5 2

0.5

1

1.5

Pulse Phase

PIN

10−30 keV

0.5

1

1.5PIN

30−40 keV

0.5

1

1.5PIN

40−55 keV

0

1

2 PIN

55−70 keV

0 0.5 1 1.5 2

0.8

1

1.2

Pulse Phase

GSO

40−55 keV

0

0.5

1

1.5

2

GSO

55−70 keV

0

1

2GSO

70−100 keV

0

1

2 GSO 100−120

keV

0 0.5 1 1.5 2

0

2

Pulse Phase

GSO 120−150 keV

Fig. 2 Energy-resolved pulse profiles of KS 1947+300 obtained from XIS-0, HXD/PIN and HXD/GSO light curves at various energy

ranges obtained from the first Suzaku observation of the pulsar on 2013 October 22. The error bars represent 1σ uncertainties. Two

pulses in each panel are shown for clarity.

0.5

1

1.5

XIS−00.5−5 keV

0.5

1

1.5

XIS−05−7 keV

0.5

1

1.5

XIS−07−10 keV

Norm

aliz

ed I

nte

nsi

ty

0 0.5 1 1.5 2

0.5

1

1.5

Pulse Phase

PIN10−30 keV

0.5

1

1.5

PIN

30−40 keV

0.5

1

1.5

PIN40−55 keV

0

0.5

1

1.5

2

PIN

55−70 keV

0 0.5 1 1.5 20.6

0.8

1

1.2

Pulse Phase

GSO40−55 keV

0

0.5

1

1.5 GSOkeV

55−70

0

0.5

1

1.5

2

GSO 70−100

keV

0

1

2GSO 100−120

keV

0 0.5 1 1.5 2

0

2

Pulse Phase

GSO 120−150 keV

Fig. 3 Energy-resolved pulse profiles of KS 1947+300 obtained from XIS-0, HXD/PIN and HXD/GSO light curves at various energy

ranges, during the second Suzaku observation. The presence of a dip-like feature can be clearly seen in the pulse profiles. The error bars

represent 1σ uncertainties. Two pulses in each panel are shown for clarity.

XIS-0, XIS-1, XIS-3, PIN and GSO detectors were used

in our analysis. The procedure followed to extract source

and background spectra was described in the earlier

section. To improve statistics, we re-binned the source

spectra obtained from XISs and PIN event data to have a

minimum of 20 counts per energy channel. However, GSO

spectra were grouped as suggested by the instrument team.

Like other bright X-ray sources where a systematic error

of up to 3% was added to XIS spectra (Cyg X-1; Nowak

et al. 2011), a systematic error of 1% was added to XIS

spectra of KS 1947+300 for the cross calibration issues

between back and front illuminated CCDs. Simultaneous

spectral fitting was carried out in the ∼1–110 keV range

for both observations by using the XSPEC (version 12.8.2)

package. During spectral fitting, data in the ranges of

1.7–1.9 keV and 2.2–2.4 keV were ignored due to the

presence of known Si and Au edges in XISs spectra. All

the model parameters were tied together except the values

of relative normalization of detectors which were kept free

during simultaneous spectral fitting.

We used a high-energy cut-off power law, a cut-off

power law and the Negative and Positive power law with

EXponential cutoff (NPEX) model to describe the pulsar

continuum. We found that all three models can explain the

continuum spectrum well. Along with the interstellar ab-

sorption, a blackbody component for the soft X-ray ex-

cess and a Gaussian function for iron emission were re-

quired to fit the spectra. Though the broad-band spectral

fitting yielded an emission line at ∼6.5 keV with a width

of ∼0.2 keV, careful investigations of the residuals near the

line energy confirmed the presence of two iron emission

lines at 6.4 and 6.7 keV. Therefore, we added Gaussian

functions at 6.4 and 6.7 keV in our broad-band spectral

fitting. We identified these lines as emission from neu-

tral and He-like iron atoms. It was found that the addition

of a partial covering component to the above continuum

models improved the spectral fitting further with signifi-

cant improvement in the χ2 values (∆χ2 ≥ 70). This com-

ponent has been used to investigate the cause of absorp-

tion dips at certain phases of the pulse profiles of BeXB

Suzaku Observations of KS 1947+300 77–5

0.01

0.1

keV

(P

hoto

ns

cm−

2 s

−1 k

eV−

1)

1 10 100

−202

χ

Energy (keV)

Fig. 4 Broad-band (1–110 keV energy range) spectrum of

KS 1947+300 obtained with the XIS-0, XIS-1, XIS-3, PIN and

GSO detectors of the first Suzaku observation during the October

2013 outburst along with the best-fit model comprising a partially

absorbed NPEX continuum model, a blackbody component for

soft X-ray excess, a Gaussian function for the iron emission line

and fixed cyclotron absorption component. The contributions of

the residuals to the χ2 for each energy bin for the best-fit model

are shown in the bottom panel.

pulsars (Paul & Naik 2011). This component, therefore,

was used to probe the cause of the observed absorption

dip in the pulse profiles of KS 1947+300. A cyclotron

line at 12.2 keV that was detected from NuSTAR observa-

tions was also included in the spectral model. Since Suzaku

data cannot constrain the line region well, in our analysis,

we fixed the cyclotron line parameters, i.e. line energy at

12.2 keV, width at 2.5 keV and depth at 0.16 as obtained

from NuSTAR observations (Furst et al. 2014). Among the

three continuum models, the partial covering NPEX model

along with other spectral components was found to be the

best-fit model for both Suzaku observations.

Best-fitted spectral parameters obtained from all three

models are given in Table 2 for both observations. The

energy spectra for the partial covering NPEX continuum

model along with residuals are shown in Figures 4 and 5

for the first and second Suzaku observations, respectively.

The values of additional absorption column density (NH2)

were found to be significantly higher than the values of

Galactic absorption column density (NH1) (Table 2). The

pulsar spectrum was marginally hard at the peak of the

outburst, i.e. during the second observation compared to

the first observation. The soft excess component was found

to be stronger during the second observation (peak of the

outburst) with higher values of blackbody temperature and

flux compared to those during the first observation.

4.2 Pulse-Phase Resolved Spectroscopy

To investigate the nature of the absorption dip in hard X-

ray pulse profiles, as well as the nature of the soft ex-

cess component and the evolution of other spectral parame-

ters during both Suzaku observations, pulse-phase resolved

spectroscopy was carried out by accumulating source spec-

10−3

0.01

0.1

1

keV

(P

hoto

ns

cm−

2 s

−1 k

eV−

1)

1 10 100

−202

χ

Energy (keV)

Fig. 5 1–110 keV energy spectrum of KS 1947+300 obtained

with the XIS-0, XIS-1, XIS-3, PIN and GSO detectors of the sec-

ond Suzaku observation during the October 2013 outburst along

with the best-fit model comprised of a partially absorbed NPEX

continuum model, a blackbody component for soft X-ray excess,

a Gaussian function for the iron emission line and fixed cyclotron

absorption component. The contributions of the residuals to the

χ2 for each energy bin for the best-fit model are shown in the

bottom panel.

tra from XISs, PIN and GSO detectors in 9 and 10 phase

bins for the first and second observations, respectively.

Background spectra, response matrices and effective area

files used in phase-averaged spectroscopy were also used

in the phase-resolved spectroscopy. Simultaneous spectral

fitting was carried out for phase-resolved spectra obtained

from both observations by using a partial covering NPEX

continuum model along with other components. During

fitting, the equivalent Galactic hydrogen column density

(NH1 ; expected to be constant along the source direction),

energy and width of iron emission lines, cyclotron line pa-

rameters (line energy, width and depth) and relative instru-

ment normalizations were fixed at corresponding phase-

averaged values. Due to a lack of sufficient photons, the

iron emission lines were not resolved during the phase-

resolved spectral fitting. It was found that the change in

spectral parameters over the pulse phase are similar for

both the observations and are shown in Figures 6 and 7

for the first and second observations, respectively. Pulse

profiles of the pulsar obtained from XIS-0 and PIN light

curves of both observations are shown in the top two pan-

els of the left and right panels of Figures 6 and 7. Changes

in the spectral parameters such as additional column den-

sity (NH2), covering fraction, blackbody temperature for

soft X-ray excess and soft X-ray excess flux with pulse

phases are shown in subsequent panels on the left sides of

Figures 6 and 7. The panels in the right side of Figures 6

and 7 show the changes in soft (1–10 keV) and hard X-ray

(10–100 keV) fluxes, power-law photon index and cutoff

energy.

All the spectral parameters plotted in Figures 6 and

7 were found to be variable with pulse phase of the pul-

sar. Additional column density was found to be marginally

higher at the phase of the absorption dip in the hard X-

77–6 P. Epili, S. Naik & G. K. Jaisawal

0.5

1

1.5

XIS

0.5

1

1.5

PIN

Norm

aliz

ed I

nte

nsi

ty

4

6

8

NH

2

(10

22 u

nit

s)

0.4

0.6

0.8

Cover

ing

frac

tion

0.4

0.5

0.6

kBT

(keV

)

0 0.5 1 1.5 2

0.2

0.4

0.6

Pulse Phase

(10

−9 u

nit

s)

BB

Flu

x

0.5

1

1.5

XIS

0.5

1

1.5

Norm

aliz

ed I

nte

nsi

ty

PIN

1

2

3

(10

−9 u

nit

s)

1−10.0 keV

Flu

x2

4

6

8

Flu

x

(10

−9 u

nit

s)10−100 keV

1

1.5

Γ

0 0.5 1 1.5 2

10

20

30

40

Pulse Phase

Hig

h E

cut

(keV

)

Fig. 6 Spectral parameters obtained from the phase-resolved

spectroscopy of KS 1947+300 during the first Suzaku observa-

tion in October 2013. The first and second panels on both sides

show pulse profiles of the pulsar in the 0.5–10 keV (XIS-0) and

10–70 keV (HXD/PIN) energy ranges. The values of NH2 , cov-

ering fraction, blackbody temperature and blackbody flux are

shown in the third, fourth, fifth and sixth panels in the left side

of the figure, respectively. The soft X-ray flux in the 1–10 keV

range, hard X-ray flux in the 10–100 keV range, photon index and

high energy cutoff are shown in the third, fourth, fifth and sixth

panels in the right side of the figure, respectively. The blackbody

flux and source fluxes in 1–10 and 10–100 keV are quoted in the

units of 10−9 erg cm−2 s−1. The errors in the figure are estimated

at the 90% confidence level.

ray pulse profile. Blackbody temperature, blackbody flux

and source flux in the 1–10 keV range showed a similar

variation pattern as the soft X-ray pulse profile over pulse

phases. This confirmed the pulsating nature of the soft X-

ray excess component in phase with the source flux. The

values of power-law photon index and high energy cutoff

were found to be higher during the main dip phases. The

dependence of several spectral parameters on additional

column density and source flux in the 1–10 keV range

were investigated and are shown in Figures 8 and 9. The

left panels show the dependence of power-law photon in-

dex and blackbody temperature on the additional column

density whereas the right panels show the dependence of

blackbody temperature and blackbody flux on the source

flux in the 1–10 keV range. It was found that the black-

body temperature and flux showed a positive correlation

with the soft X-ray flux in the 1–10 kev range. The value

of power-law photon index was found to be high at a low

value of additional column density (NH2) and decreased

with the increase in NH2 .

5 DISCUSSION AND CONCLUSIONS

5.1 Spin-Period and Magnetic Field of the Pulsar

KS 1947+300 was observed with Suzaku at two epochs

during its 2013 X-ray outburst. Although the observations

were only a month apart, estimated spin periods of the pul-

sar during both the observations were found to be differ-

ent. The spin period during the second observation was

0.5

1

1.5 XIS

0.5

1

1.5 PIN

Norm

aliz

ed I

nte

nsi

ty

5

10

15

20

NH

2

(10

22 u

nit

s)

0.2

0.4

0.6

Cover

ing

frac

tion

0.5

0.6

0.7

kBT

(keV

)

0 0.5 1 1.5 2

0.5

1

Pulse Phase

(10

−9 u

nit

s)

BB

Flu

x

0.5

1

1.5 XIS

0.5

1

1.5

Norm

aliz

ed I

nte

nsi

ty

PIN

2

4

(10

−9 u

nit

s)

1−10.0 keV

Flu

x

5

10

Flu

x

(10

−9 u

nit

s)

10−100 keV

0.5

1

1.5

Γ

0 0.5 1 1.5 2

10

15

20

25

Pulse Phase

Hig

h E

cut

(keV

)

Fig. 7 Spectral parameters obtained from the phase-resolved

spectroscopy of KS 1947+300 during the second Suzaku obser-

vation in November 2013. The first and second panels in both

sides show pulse profiles of the pulsar in the 0.5–10 keV (XIS-0)

and 10–70 keV (HXD/PIN) energy ranges. The values of NH2 ,

covering fraction, blackbody temperature and blackbody flux are

shown in the third, fourth, fifth and sixth panels in the left side

of the figure, respectively. The soft X-ray flux in the 1–10 keV

range, hard X-ray flux in the 10–100 keV range, photon index and

high energy cutoff are shown in the third, fourth, fifth and sixth

panels in the right side of the figure, respectively. The blackbody

flux and source fluxes in 1–10 and 10–100 keV are expressed in

units of 10−9 erg cm−2 s−1. The errors in the figure are estimated

at the 90% confidence level.

4 6 8 100.5

1

1.5

Pow

er l

aw I

ndex

( Γ

)

4 6 8 10

0.4

0.5

0.6

BB

−T

emp. (k

BT

in k

eV)

NH2

(in 1022 atoms cm−2 )

1 2 3

0.4

0.5

0.6

BB

−T

emp. (k

BT

in k

eV)

1 2 3

0.2

0.4

0.6

BB

−F

lux (

in 0

.5−

6 k

eV)

Flux (in 1−10 keV)

Fig. 8 Dependence of different spectral parameters obtained from

the phase-resolved spectroscopy of KS 1947+300 during the first

Suzaku observation. The blackbody flux and 1–10 keV source

flux are quoted in the units of 10−9 erg cm−2 s−1.

smaller than the first observation, indicating the pulsar was

spinning up during the X-ray outburst. While comparing

the recent measurements of the spin period of the pul-

sar with what was reported from observations with several

other observatories (as quoted in Table 1), it was found that

the pulsar was continuously spinning up during the entire

2013 X-ray outburst. During X-ray outbursts, spinning-up

of the neutron star is expected due to transfer of angular

momentum from accreting matter at the magnetic poles.

Ghosh & Lamb (1979) formulated the dependence of the

spin-up rate of a pulsar on its luminosity as P ∝ L6/7.

Suzaku Observations of KS 1947+300 77–7

Table 2 Best-fitting spectral parameters (with 90% errors) obtained from two Suzaku observations of KS 1947+300 during the 2013

outburst. Model-1: partial covering NPEX model with Gaussian components and a cyclotron absorption line; Model-2: partial covering

high-energy cutoff model with Gaussian components and a cyclotron absorption line; Model-3: partial covering cutoff power law

model with Gaussian components and a cyclotron absorption line. The cyclotron line parameters were fixed at the values from Furst

et al. (2014).

Parameter October 2013 (Obs.1) November 2013 (Obs.2)

Model-1 Model-2 Model-3 Model-1 Model-2 Model-3

NH1a 0.50±0.02 0.50±0.02 0.52±0.02 0.48±0.02 0.50±0.02 0.53±0.02

NH2b 7.6±1.0 8.2±1.1 7.6±1.1 11.3±2.5 10.7±1.7 12.1±4.8

Covering fraction 0.44±0.06 0.45±0.06 0.43±0.06 0.35±0.06 0.35±0.06 0.27±0.07

Photon Index (Γ) 0.67±0.03 0.95±0.04 0.92±0.05 0.62±0.04 0.93±0.04 0.93±0.03

Ecut (keV) 10.2±0.3 5.4±0.4 20.2±0.9 10.6±0.4 5.9±0.3 21.6±0.7

Efold (keV) – 21.0±0.7 – – 21.6±0.7 –

Blackbody temp. kT (keV) 0.54±0.02 0.56±0.02 0.54±0.02 0.63±0.03 0.65±0.02 0.65±0.03

Blackbody fluxc 0.88±0.13 0.96±0.13 0.77±0.13 1.34±0.19 1.40±0.20 1.02±0.19

Emission lines

Fe Kα line energy (keV) 6.42±0.03 6.42±0.03 6.42±0.03 6.45±0.02 6.45±0.02 6.45±0.02

Width of Fe line (keV) 0.01+0.06−0.01 0.01+0.06

−0.01 0.01+0.06−0.01 0.01+0.04

−0.01 0.01+0.04−0.01 0.01+0.04

−0.01

Eq. width of Fe line (eV) 18±2 19±2 19±2 20±3 20±3 22±2

Line energy (keV) 6.66±0.05 6.66±0.05 6.66±0.05 6.71±0.04 6.71±0.04 6.71±0.02

Line width (keV) 0.01+0.07−0.01 0.01+0.07

−0.01 0.01+0.07−0.01 0.01+0.07

−0.01 0.01+0.07−0.01 0.01+0.07

−0.01

Equivalent width (eV) 17±2 18±2 18±2 11±3 11±3 13±3

Source flux

Fluxc (1–10 keV) 2.6±0.2 2.7±0.2 2.6±0.2 4.3±0.4 4.3±0.4 4.1±0.4

Fluxc (10–70 keV) 5.4±0.5 5.3±0.4 5.3±0.3 6.6±0.6 6.5±0.5 6.5±0.4

Fluxc (70–100 keV) 0.21±0.03 0.27±0.02 0.27±0.02 0.38±0.03 0.38±0.03 0.38±0.02

Reduced χ2 1.18 (942) 1.25 (970) 1.25 (971) 1.08 (970) 1.10 (942) 1.13 (971)

Notes: a: Equivalent hydrogen column density (in 1022 atom cm−2 units); b: Additional hydrogen column density (in 1022 atom cm−2 units);c: Absorption corrected flux in units of 10−9 erg cm−2 s−1.

5 10 15

0.6

0.8

1

1.2

Po

wer

law

In

dex

( Γ

)

5 10 150.4

0.5

0.6

0.7

BB

−T

emp

. (k

BT

in

keV

)

NH2

(in 1022 atoms cm−2 )

1 2 3 4 5

0.5

0.6

0.7

BB

−T

emp

. (k

BT

in

keV

)

1 2 3 4 5

0.5

1

BB

−F

lux

(in

0.5

−6

keV

)

Flux (in 1−10 keV)

Fig. 9 Dependence of different spectral parameters obtained from

the phase-resolved spectroscopy of KS 1947+300 during the sec-

ond Suzaku observation. The blackbody flux and 1–10 keV source

flux are quoted in units of 10−9 erg cm−2 s−1.

In the present work, the pulsar spin period was found to

be 18.7878 s (Suzaku) at the peak of the outburst which

decreased to 18.77088 s (NuSTAR observation) during the

decay of the outburst (see Table 1). The observed spin-up

of KS 1947+300 can be attributed to the change in angular

momentum due to the torque exerted by accreting matter

on the neutron star. A similar type of rapid spin-up was

also observed during the declining phase of the 2001 out-

burst of the pulsar (Naik et al. 2006).

The pulsar was showing a spin-up trend during Suzaku

and NuSTAR observations. The observed spin-up (angular

frequency) rate (ωsu) can be used to estimate the magnetic

field of the pulsar by considering the quasi-spherical set-

tling accretion theory (Shakura et al. 2012; Postnov et al.

2015). According to this theory,

ωsu ≃ 10−9[Hz d−1] Πsu µ1/1130 v−4

8

(Pb

10 d

)−1

M7/1116 ,

(1)

where ωsu is spin-up rate which is estimated to be

1.27×10−5[Hz d−1] (present case) and Πsu is the di-

mensionless parameter from settling accretion theory. The

value of Πsu is independent of the system and is in

the range of ∼4.6 to 10 (Shakura et al. 2012; Postnov

et al. 2015). In the present case, Πsu was assumed to

be 4.6. The dipole magnetic moment of the neutron star

µ30=µ/1030[G cm3] and is related to the magnetic field (B)

77–8 P. Epili, S. Naik & G. K. Jaisawal

through the relation µ = BR3/2 (R is the neutron star ra-

dius, assumed to be 10 km). Stellar wind velocity v8 =v/108 [cm s−1] is considered to be 200 km s−1 for typ-

ical BeXBs (Waters et al. 1988). The mass accretion rate

M16=M /1016[g s−1] for the luminosity of 1038 erg s−1

was estimated to be M16 = 100 during Suzaku ob-

servations. The orbital period Pb of KS 1947+300 is

40.42 d (Galloway et al. 2004). Using the above param-

eters, the magnetic field of the pulsar was estimated to be

∼1.2×1012 G. The estimated value of the magnetic field

by using the observed spin-up rate in KS 1947+300 was

found to agree with that obtained from the detection of the

cyclotron absorption line at 12.2 keV.

5.2 Pulse Profiles

In the present work, the pulse profile of KS 1947+300

was found to be simple at lower energies. As the energy

increases, a dip like structure appears in the pulse pro-

file and is detected up to 70 keV. The depth of the dip

is found to increase with energy and is maximum in the

30–40 keV range. Such type of behavior is not generally

seen in pulse profiles of other BeXB pulsars. We tried to

investigate the cause of the absorption dip in pulse pro-

files at hard X-rays (>10 keV) through phase-resolved

spectroscopy. A marginal enhancement in the additional

column density at the dip phase was detected. Such a

low value (≤20×1022 cm−2) of additional column den-

sity, however, cannot absorb the hard X-ray photons up

to ∼70 keV. KS 1947+300 was also observed at differ-

ent luminosity levels with several observatories such as

BeppoSAX, RXTE, INTEGRAL and NuSTAR. The pulse

profiles obtained from these observations were found to

be similar to that obtained from Suzaku observations. The

dip was only seen in hard X-rays (Galloway et al. 2004;

Tsygankov & Lutovinov 2005; Naik et al. 2006; Furst et al.

2014). Therefore, the presence of the dip in hard X-ray

pulse profiles of KS 1947+300 is possibly intrinsic to the

pulsar.

In general, the pulse profiles of BeXB pulsars are seen

to be strongly energy and luminosity dependent. Single

or multiple absorption dips, prominent at soft X-ray, are

seen in the pulse profiles of these pulsars (Paul and Naik

2011 and references therein). With an increase in energy,

the depth of the dip decreases and becomes invisible at

higher energies. It is widely believed that these dips in the

pulse profile are due to obscuration/absorption of soft X-

ray photons by matter present close to the neutron star.

In some cases, single or multiple dips were observed at

high energies, e.g. up to 70 keV in pulse profiles of EXO

2030+275 (Naik et al. 2013; Naik & Jaisawal 2015). The

presence of additional dense matter at dip phases was de-

tected from phase-resolved spectroscopy and was inter-

preted as the cause of absorption dips in the pulse profiles

of EXO 2030+375. In KS 1947+300 (present work), how-

ever, the origin of the dip in hard X-ray pulse profiles (up

to ∼70 keV) is not due to the presence of additional matter

at a certain phase of the pulsar.

The pulse profile of the pulsars can be affected by cy-

clotron resonance scattering and geometrical effects. These

effects can play a vital role and shape the anomaly or dip

in the pulse profiles. In KS 1947+300, a cyclotron absorp-

tion line was detected at ∼12.2 keV (Furst et al. 2014).

The beam function of an accreting pulsar can be affected

by the presence of strong cyclotron resonance scattering

which can produce a significant change in the pulse pro-

file, e.g. phase-shift (lag) (Schonherr et al. 2014). Similar

effects were detected in BeXB pulsars such as V 0332+53

(Tsygankov et al. 2006), 4U 0115+63 (Ferrigno et al. 2011)

and GX 304–1 (Jaisawal et al. 2016). However, this is not

the case in KS 1947+300 as the strength of the observed

dip increased with energy and became prominent in the

∼30–40 keV energy range. Around this energy, however,

the influence of cyclotron resonance scattering is not as ef-

fective as compared to energies closer to ∼12 keV. In addi-

tion, any significant change in the pulse profiles (beam pat-

tern) or phase-lags was not observed in the energy resolved

pulse profiles (Figs. 2 and 3). Therefore, we expect that the

cyclotron scattering is not causing the hard X-ray dip in the

pulse profiles of KS 1947+300. Alternatively, the presence

of a single dip in hard X-ray profiles suggests the direct

viewing of the pole of the neutron star through the accre-

tion column. At such high luminosity (∼1038 erg s−1) like

the one found by the Suzaku observations of KS 1947+300,

a radiation pressure dominated shock is expected to form

above the surface of the neutron star which can absorb the

photons up to higher energies.

In this case, the position of the absorption dip should

be at the peak of the pulse profiles. However, the asymmet-

ric phase position of the dip with respect to the main dip

in pulse profiles (Figs. 2 and 3) discards the hypothesis of

direct viewing of the pulsar along the magnetic axis. It is

accepted that the pulse profile of X-ray pulsars depends on

the geometry and viewing angle of the emission region or

accretion column (Kraus et al. 1995; Caballero et al. 2011;

Sasaki et al. 2012). We suggest that the dip in the hard

X-ray pulse profiles of KS 1947+300 is due to these geo-

metrical effects. The dip was absent in the soft X-ray pulse

profiles. The presence of strong soft X-ray excess (which

was found pulsating in phase with the neutron star) may

cancel the effect of the absorption dip, producing single

pulse profiles in soft X-ray bands.

5.3 Spectroscopy

In this paper, broad-band phase-averaged and phase-

resolved spectra of KS 1947+300 are presented by using

two Suzaku observations of the 2013 giant outburst. During

both observations, the values of estimated Galactic col-

umn density were comparable. However, the values of the

additional column density were found to be significantly

higher than the Galactic value. The higher values of ad-

ditional absorption column density indicate the presence

Suzaku Observations of KS 1947+300 77–9

of additional matter near the neutron star during the X-ray

outburst. During both the observations, a soft X-ray ex-

cess was clearly detected and its temperature was found to

be high at the peak of the outburst (second observation).

Assuming the blackbody emitting region to be spherically

symmetric, the radius of the soft X-ray excess emitting re-

gion in KS 1947+300 is estimated to be ∼29–31 km. It

implies that the soft X-ray excess emitting region is close

to the neutron star surface. The pulsating nature of the soft

X-ray excess in KS 1947+300 agrees with the above ar-

gument. Therefore, the accretion column and/or accretion

streams are the most probable origin site of soft X-ray ex-

cess emission in KS 1947+300 (Naik & Paul 2002; Naik

& Paul 2004; Hickox et al. 2004).

Apart from the detection of soft X-ray excess and the

presence of additional matter around the pulsar, a change

in the power-law photon index and high energy cutoff with

pulse phase was seen in KS 1947+300. A similar type

of variation was also seen in other BeXB pulsars such as

EXO 2030+275 (Naik et al. 2013; Naik & Jaisawal 2015).

Narrow iron Kα and He-like iron emission lines at ∼6.4

and 6.7 keV were detected during both the observations.

During BeppoSAX observations of KS 1947+300 in the

2001 outburst, an emission line at 6.7 keV was detected

whereas the iron Kα line was absent in the pulsar spec-

tra (Naik et al. 2006). The 6.7 keV line was identified

as the emission feature from helium-like iron atoms. A

cyclotron absorption feature at ∼12.2 keV was detected

in KS 1947+300 from NuSTAR observations during the

2013 outburst (Furst et al. 2014). Detection of the cy-

clotron line is a unique tool to directly estimate the mag-

netic field of the pulsar by using the 12-B-12 rule or

Ecyc = 11.6B12×(1+z)−1. Using the detected cyclotron

absorption line at 12.2 keV, the strength of the surface

magnetic field was estimated to be ∼1.1×1012(1 + z) G

(Furst et al. 2014). Though the pulsar was observed with

NuSTAR at three epochs, the cyclotron line was only de-

tected during the second observation and there was no sig-

nature of the presence of its harmonics in the pulsar spec-

trum. We also did not find harmonics of the 12.2 keV cy-

clotron line in pulsar spectra obtained from Suzaku ob-

servations. There are several pulsars where a fundamen-

tal cyclotron line is seen in the broad-band spectra without

the detection of its harmonics, e.g. Cen X-3 (Suchy et al.

2008; Naik et al. 2011), Swift J1626.6–5156 (DeCesar

et al. 2013), and IGR J17544-2619 (Bhalerao et al. 2015).

The 1–100 keV luminosity of the pulsar was estimated to

be ∼9.8×1037 and 1.3×1038 erg s−1 during the first and

second Suzaku observations, respectively. Critical lumi-

nosity was calculated to investigate the luminosity regime

of the pulsar by assuming parameters of a canonical neu-

tron star with cyclotron line energy at 12.2 keV in the re-

lation of Becker et al. (2012). This was estimated to be

∼1.6×1037 erg s−1. It is clear that the pulsar was accreting

in the super-Eddington regime (above the critical luminos-

ity) during the October 2013 (present work) and November

2000 (Naik et al. 2006) outbursts.

In summary, we reported on the timing and broad-band

spectral properties of the pulsar KS 1947+300 by using

Suzaku observations taken during the 2013 outburst. Soft

X-ray pulse profiles were found to be smooth and sin-

gle peaked. However, hard X-ray pulse profiles showed

the presence of an absorption dip like feature. The 1–

110 keV broad-band spectrum of the pulsar was described

with a partially absorbed NPEX continuum model along

with a blackbody component. Phase-resolved spectroscopy

revealed marginal enhancement in the additional column

density at the dip phase, which suggests that the dip is not

because of the absorption of hard X-ray photons. Another

mechanism such as a geometrical effect could be a possi-

ble cause for the presence of a dip in the hard X-ray pulse

profiles of KS 1947+300. Detection of pulsation in the

soft X-ray excess flux confirmed that the emitting region

is close to the neutron star, e.g. near the accretion column.

The presence of soft X-ray excess may be the cause of the

absence of the dip in soft X-ray profiles. We estimated the

magnetic field of the pulsar by using the observed spin-up

rate during Suzaku and NuSTAR observations. The value

was found to be 1.2×1012 G and comparable to that ob-

tained from the cyclotron line energy.

Acknowledgements We sincerely thank the referee for

his valuable comments and suggestions which improved

the paper significantly. The research work at the Physical

Research Laboratory is funded by the Department of

Space, Government of India. The authors would like

to thank all the members of the Suzaku mission for

their contributions in the instrument preparation, space-

craft operation, software development and in-orbit in-

strumental calibration. This research has made use of

data obtained through HEASARC Online Service, pro-

vided by NASA/GSFC, in support of NASA High Energy

Astrophysics Programs.

References

Becker, P. A., Klochkov, D., Schonherr, G., et al. 2012, A&A,

544, A123

Bhalerao, V., Romano, P., Tomsick, J., et al. 2015, MNRAS, 447,

2274

Borozdin, K., Gilfanov, M., Sunyaev, R., et al. 1990, Soviet

Astronomy Letters, 16, 345

Caballero, I., Kraus, U., Santangelo, A., Sasaki, M., &

Kretschmar, P. 2011, A&A, 526, A131

Caballero, I., & Wilms, J. 2012, Mem. Soc. Astron. Italiana, 83,

230

Chakrabarty, D., Koh, T., Bildsten, L., et al. 1995, ApJ, 446, 826

DeCesar, M. E., Boyd, P. T., Pottschmidt, K., et al. 2013, ApJ,

762, 61

Ferrigno, C., Falanga, M., Bozzo, E., et al. 2011, A&A, 532, A76

Finger, M. H., Jenke, P. A., & Wilson-Hodge, C. A. 2015, The

Astronomer’s Telegram, 7017

77–10 P. Epili, S. Naik & G. K. Jaisawal

Furst, F., Pottschmidt, K., Wilms, J., et al. 2014, ApJ, 784, L40

Galloway, D. K., Morgan, E. H., & Levine, A. M. 2004, ApJ,

613, 1164

Ghosh, P., & Lamb, F. K. 1979, ApJ, 234, 296

Hickox, R. C., Narayan, R., & Kallman, T. R. 2004, ApJ, 614,

881

Jaisawal, G. K., Naik, S., & Epili, P. 2016, MNRAS, 457, 2749

James, M., Paul, B., Devasia, J., & Indulekha, K. 2010, MNRAS,

407, 285

Koyama, K., Tsunemi, H., Dotani, T., et al. 2007, PASJ, 59, 23

Kraus, U., Nollert, H.-P., Ruder, H., & Riffert, H. 1995, ApJ, 450,

763

Levine, A., & Corbet, R. 2000, IAU Circ., 7523

Mitsuda, K., Bautz, M., Inoue, H., et al. 2007, PASJ, 59, 1

Naik, S., & Paul, B. 2002, Journal of Astrophysics and

Astronomy, 23, 27

Naik, S., & Paul, B. 2004, ApJ, 600, 351

Naik, S., Callanan, P. J., Paul, B., & Dotani, T. 2006, ApJ, 647,

1293

Naik, S., Paul, B., & Ali, Z. 2011, ApJ, 737, 79

Naik, S., Maitra, C., Jaisawal, G. K., & Paul, B. 2013, ApJ, 764,

158

Naik, S., & Jaisawal, G. K. 2015, RAA (Research in Astronomy

and Astrophysics), 15, 537

Negueruela, I., Israel, G. L., Marco, A., Norton, A. J., & Speziali,

R. 2003, A&A, 397, 739

Nowak, M. A., Hanke, M., Trowbridge, S. N., et al. 2011, ApJ,

728, 13

Okazaki, A. T., & Negueruela, I. 2001, A&A, 377, 161

Paul, B., & Naik, S. 2011, Bulletin of the Astronomical Society

of India, 39, 429

Postnov, K. A., Mironov, A. I., Lutovinov, A. A., et al. 2015,

MNRAS, 446, 1013

Reig, P. 2011, Ap&SS, 332, 1

Sasaki, M., Muller, D., Kraus, U., Ferrigno, C., & Santangelo, A.

2012, A&A, 540, A35

Schonherr, G., Schwarm, F.-W., Falkner, S., et al. 2014, A&A,

564, L8

Serlemitsos, P. J., Soong, Y., Chan, K.-W., et al. 2007, PASJ, 59,

9

Shakura, N., Postnov, K., Kochetkova, A., & Hjalmarsdotter, L.

2012, MNRAS, 420, 216

Skinner, G. K. 1989, IAU Circ., 4850

Suchy, S., Pottschmidt, K., Wilms, J., et al. 2008, ApJ, 675, 1487

Swank, J., & Morgan, E. 2000, IAU Circ., 7531

Takahashi, T., Abe, K., Endo, M., et al. 2007, PASJ, 59, 35

Tsygankov, S. S., & Lutovinov, A. A. 2005, Astronomy Letters,

31, 88

Tsygankov, S. S., Lutovinov, A. A., Churazov, E. M., &

Sunyaev, R. A. 2006, MNRAS, 371, 19

Waters, L. B. F. M., van den Heuvel, E. P. J., Taylor, A. R.,

Habets, G. M. H. J., & Persi, P. 1988, A&A, 198, 200

MNRAS 472, 3455–3466 (2017) doi:10.1093/mnras/stx2247Advance Access publication 2017 September 5

Decade long RXTE monitoring observations of Be/X-ray binary pulsarEXO 2030+375

Prahlad Epili,‹ Sachindra Naik,‹ Gaurava K. Jaisawal‹ and Shivangi GuptaAstronomy and Astrophysics Division, Physical Research Laboratory, Navrangapura, Ahmedabad 380009, Gujarat, India

Accepted 2017 August 30. Received 2017 August 29; in original form 2017 May 13

ABSTRACTWe present a comprehensive timing and spectral studies of Be/X-ray binary pulsar EXO2030+375 using extensive Rossi X-ray Timing Explorer observations from 1995 till 2011,covering numerous Type I and 2006 Type II outbursts. Pulse profiles of the pulsar were foundto be strongly luminosity dependent. At low luminosity, the pulse profile consisted of a mainpeak and a minor peak that evolved into a broad structure at high luminosity with a significantphase shift. A narrow and sharp absorption dip, also dependent on energy and luminosity,was detected in the pulse profile. Comparison of pulse profiles showed that the featuresat a particular luminosity are independent of type of X-ray outbursts. This indicates that theemission geometry is solely a function of mass accretion rate. The broad-band energy spectrumwas described with a partial covering high energy cutoff model as well as a physical modelbased on thermal and bulk Comptonization in accretion column. We did not find any signatureof cyclotron resonance scattering feature in the spectra obtained from all the observations.A detailed analysis of spectral parameters showed that, depending on source luminosity,the power-law photon index was distributed in three distinct regions. It suggests the phasesof spectral transition from sub-critical to super-critical regimes in the pulsar as proposedtheoretically. A region with constant photon index was also observed in ∼(2–4) × 1037 erg s−1

range, indicating critical luminosity regime in EXO 2030+375.

Key words: stars: neutron – pulsars: individual: EXO 2030+375 – X-rays: stars.

1 IN T RO D U C T I O N

Most of the accretion powered binary X-ray pulsars are amongthe brightest sources in our Galaxy. They belong to the class ofhigh mass X-ray binaries in which a neutron star accretes matterfrom a massive >10 M� main-sequence companion. Dependingon the evolutionary state of the donor, mass transfer from the com-panion star to the compact object takes place through capture ofstellar wind or accretion from a huge circumstellar disc around thecompanion star (Paul & Naik 2011). Among confirmed high massX-ray binaries (HMXBs), Be/X-ray binaries (BeXBs) representabout two-third population with neutron star as the compact ob-ject. The optical companion in these systems is a non-supergiant Bor O type star that shows emission lines in its spectrum (Reig 2011).An excess emission in the infrared band is also observed from thesecompanion stars in BeXBs. It is believed that due to rapid rotation,the Be star expels material equatorially forming a huge disc, calledas circumstellar disc, around it. The observed emission lines andinfrared excess in the spectrum are attributed to the presence ofcircumstellar disc around the central Be star.

� E-mail: prahlad@prl.res.in (PE); snaik@prl.res.in (SN); gaurava@prl.res.in (GKJ)

A neutron star in BeXBs revolves in a wide and moderate ec-centric orbit. While passing close to the periastron, an abrupt massaccretion from the circumstellar envelope on to the neutron stargives rise to strong X-ray outbursts. The intensity during such out-bursts increases up to an order of magnitude than the quiescentphase. BeXBs generally show periodic or normal (Type I) X-rayoutbursts that occur at the periastron passage of the neutron star.These outbursts cover a small fraction of the orbit (<20–30 per cent)and last for a few days to weeks (Stella, White & Rosner 1986).Another class of X-ray outbursts such as giant outbursts (Type II)are also seen from the neutron stars in BeXBs. These outburstscover a significant fraction or multiple orbits lasting for severalweeks to months. Type II X-ray outbursts are quite rare and inde-pendent of the orbital phase or periastron passage of the binary.During normal and giant X-ray outbursts, the luminosity of thepulsar generally reaches up to ≤1037 and 1038 erg s−1, respectively(Okazaki & Negueruela 2001).

The spin period of BeXB pulsars ranges from a few secondsto about thousand seconds. During outbursts, change in spin pe-riod of the pulsars has also been observed. This occurs because oftorque transfer from accreting material to the neutron star. Accre-tion powered pulsars usually show broad-band emission rangingfrom soft to hard X-rays. It is interpreted as due to thermal or bulkComptonization of seed photons from the hot spots on the neutron

C© 2017 The AuthorsPublished by Oxford University Press on behalf of the Royal Astronomical Society

Downloaded from https://academic.oup.com/mnras/article-abstract/472/3/3455/4104638/Decade-long-RXTE-monitoring-observations-of-Be-Xby Physical Research Laboratory, Ahmedabad useron 17 October 2017

3456 P. Epili et al.

star surface across the accretion column (Becker & Wolff 2007).Apart from the continuum, the energy spectrum of pulsars alsoshows the presence of several other components such as soft X-rayexcess, iron fluorescence emissions, cyclotron resonance scatteringfeatures (CRSF) etc. Detection of CRSFs in the pulsar spectrumprovides a direct and most accurate method for the estimation ofsurface magnetic field of neutron stars. These features are originateddue to the resonant scattering of photons with electrons in quantizedenergy levels in the presence of strong magnetic field. Spectrum andpulse profile (representation of beam function) of the pulsars canchange depending on the mass accretion rate, accretion geometryand physical processes occurring close to the neutron star. For adetailed description on the properties of transient HMXB pulsars,refer to articles by Paul & Naik (2011) and Caballero & Wilms(2012).

Be/X-ray binary pulsar EXO 2030+375 was discovered duringa giant X-ray outburst in 1985, with EXOSAT observatory (Parmar,White & Stella 1989b). The observations during the same outburstrevealed the pulsating nature of the neutron star with a spin period of42 s along with an orbital modulation of 44.3–48.6 d. The transientnature of the pulsar was revealed during these observations. The 1–20 keV luminosity of the pulsar was observed to change by a factorof ≥2500 from quiescence over a duration of 100 d. A significantspin-up (−P/P ∼ 30 yr) of the pulsar was also observed duringthe EXOSAT observations (Parmar et al. 1989b), suggesting thepresence of an accretion disc. Using optical and infrared observa-tions, the counterpart of compact object was discovered as a highlyreddened B0 Ve star located at a distance of 7.1 kpc (Motch & Janot-Pacheco 1987; Coe et al. 1988; Wilson et al. 2002). Stollberg (1997)derived orbital parameters of the binary system by using long termBATSE monitoring data. The orbital period was determined moreprecisely and found to be 46 d. Strong luminosity dependent pulseprofile was seen during the 1985 outburst (Parmar et al. 1989a). Athigher luminosity, the pulse profile of the pulsar was characterizedby two peaks (main peak and a minor peak) separated by a phasedifference of ∼0.5. The strength of these two peaks was found toalter when the source luminosity decreased by a factor of ∼100.This observed change in strength and structure of each of the peaksin the pulse profile with luminosity was attributed to the changein the emission beam pattern e.g. from fan beam to pencil beamgeometry (Parmar et al. 1989b).

EXO 2030+375 is a unique BeXB pulsar which shows regularType I X-ray outbursts at the periastron passage (Wilson et al. 2002).At the peak of the Type I outbursts observed with the RXTE in1996–2006, source flux was approximately 100 mCrab. A corre-lation between the spin frequency and luminosity indicated thatthe pulsar was spinning-up during brighter outbursts in 1992–1994.On the other hand, a spin-down trend was observed during lowluminous outbursts in 1994–2002 (Wilson et al. 2002; Wilson,Fabregatet & Coburn 2005). EXO 2030+375 was caught into agiant outburst in 2006 June with source flux peaking up to ∼750mCrab (Krimm et al. 2006). During this giant outburst which lastedfor about 140 d, the neutron star showed a remarkable spin-up be-haviour (Wilson, Finger & Camero-Arranz 2008). After this out-burst, many intense Type I outbursts (≤300 mCrab) were detectedfor a number of orbits till the source settled to its regular mode.Since early 2015, however, the pulsar had undergone to a period oflow activity for more than a year. The peak flux drastically wentdown during this phase and hardly any X-ray enhancement was seenat the expected periastron epochs (Fuerst et al. 2016). Recent obser-vations with Swift/XRT and NuSTAR after 2016 March, however,confirmed the recurrence of X-ray activity (Type I X-ray outbursts)

along with spin-down trend in the neutron star (Kretschmar et al.2016).

The energy spectrum of pulsar obtained from the 1985 giantoutburst was described by a power-law model along with thermalblackbody component at 1.1 keV (Sun et al. 1994 and referencetherein). However, an absorbed power-law modified with high en-ergy cutoff model was widely used in later observations of EXO2030+375 during Type I and Type II outbursts (Reig & Coe 1999;Wilson et al. 2008). Apart from the 6.4 keV iron florescence emis-sion line, detection of cyclotron absorption line was reported atthree different energies such as ∼11, 36 and 63 keV in pulsar spec-tra obtained from RXTE and INTEGRAL observations during dif-ferent X-ray outbursts (Reig & Coe 1999; Klochkov et al. 2008;Wilson et al. 2008). However, Suzaku observations during 2007May–June and 2012 May Type I outbursts did not confirm the pres-ence of any such features in the pulsar spectrum (Naik et al. 2013;Naik & Jaisawal 2015). Above Suzaku observations also showedsome other interesting aspects. Along with iron lines, several emis-sion lines were also detected in the spectrum. It was the first timewhen an absorption dip was detected in the pulse profile up to as highas ∼70 keV (Naik et al. 2013). This was explained as due to the pres-ence of additional dense matter (partial absorber) at certain phasesof the pulsar. A peculiar narrow absorption dip was also detected insoft X-ray pulse profile obtained from XMM–Newton observationin 2014 May at a luminosity of ∼1036 erg s−1 (Ferrigno et al. 2016).This feature was interpreted as the effect of self-absorption fromaccretion mount on to the neutron star surface.

In this paper, we present a detailed study of decade long RXTEmonitoring observations of the pulsar over a wide range of lumi-nosity. Investigations on the pulse profiles and corresponding spec-tral parameters were performed to understand the properties of thepulsar during Type I and Type II outbursts. Along with standardcontinuum models used to describe the pulsar spectrum, we alsoused a physical model based on thermal and bulk Comptonizationof infalling plasma in the accretion column (BW model; Becker &Wolff 2007; Ferrigno et al. 2009) to describe the spectrum of EXO2030+375. We have used this model to understand the physicalproperties of accretion column across a wide range of the pul-sar luminosity. Section 2 describes the details of observations anddata analysis procedures for RXTE and NuSTAR observations. Wepresent the results obtained from timing and spectral studies inSection 3. The implication of our results are discussed in Section 4.

2 O B S E RVAT I O N S A N D A NA LY S I S

EX0 2030+375 has been observed numerous times during RXTEera i.e. between 1995–2012. These observations were performed atmultiple epochs during 2006 Type II and several Type I outbursts.We have analysed a total of 606 pointing observations for an effec-tive exposure of 1.52 million seconds in this study to understandtiming and spectral properties of the pulsar over a span of 15 years.A detail log of relevant observations are given in Table 1. A NuclearSpectroscopic Telescope Array (NuSTAR) observation of the pulsarperformed during an extended period of low activity in 2015 is alsoused in our work.

2.1 RXTE

Space based X-ray observatory RXTE was launched on 1995 De-cember 30 in a low earth orbit for understanding the physics anddynamics of compact objects. It carried highly timing and mod-erate spectral capability instruments such as Proportional Counter

MNRAS 472, 3455–3466 (2017)Downloaded from https://academic.oup.com/mnras/article-abstract/472/3/3455/4104638/Decade-long-RXTE-monitoring-observations-of-Be-Xby Physical Research Laboratory, Ahmedabad useron 17 October 2017

RXTE observations of EXO 2030+375 3457

Table 1. Log of RXTE/PCA observations of the pulsar EXO 2030+375 during Type I and Type II outbursts.

Year of observations Proposal No. of obs. Time range On source timeID (IDs) (MJD) (ksec)

1996 July P10163 18 50266.55 – 50274.56 67.421998 January P30104 2 50825.02 – 50827.77 37.942002 June P70074 21 52431.97 – 52441.47 76.912003 September P80071 15 52894.44 – 52898.32 145.872005 June – 2006 February P91089 52 53540.78 – 53776.28 145.572006 March – 2006 November P92067, P91089, P92066 143 53816.96 – 54069.97 342.342006 December – 2007 June P92422 147 54070.95 – 54279.51 221.912007 June – 2008 October P93098 79 54280.56 – 54749.53 202.412008 December – 2009 October P94098 40 54830.20 – 55114.49 92.552010 January – 2010 November P95098 43 55197.87 – 55530.08 83.792011 January – 2011 November P96098 46 55566.06 – 55895.74 106.68

array (PCA; Jahoda et al. 1996) and high energy timing experi-ment (HEXTE; Rothschild et al. 1998), sensitive in the soft to hardX-rays. The all sky monitor (ASM; Levine et al. 1996), effectivein 1.5–12 keV range was also the essential part of the observatory.The PCA detector of RXTE consisted of an array of five identi-cal Proportional Counter Units (PCUs) with a total collecting areaof ∼6500 cm2. The individual PCUs had three Xenon layers hav-ing two anode chains in each of them. Each PCU was sensitive in2–60 keV range and had a field of view of 1 deg FWHM (Jahodaet al. 2006). Including the Xenon-filled main counter, the detectoralso had a Propane ‘veto’ layer on either sides with aluminizedMylar window for background rejection. The hard X-ray unit ofRXTE, HEXTE consisted of two main clusters A and B, rock-ing orthogonally to provide simultaneous measurement of sourceand background. Each cluster was made of four NaI(Tl)/CsI(Na)phoswich scintillation detectors working in 15–250 keV range. Thetotal collecting area of both the clusters was ∼1600 cm2.

For our analysis, we have utilized publicly available RXTE ob-servations acquired from 1996 July to 2011 November carried outduring various Type I and Type II X-ray outbursts (see Table 1). Wehave mainly used the PCA standard-1 and standard-2 binned modedata for timing and spectral studies of the pulsar, respectively. Forprocessing, standard methods are followed by creating appropriategood time interval file and filter selection on all available PCUsusing HEASOFT (version 6.16) package. The source light curves wereextracted in the 2–60 keV from standard-1 data at 0.125 s time reso-lution using saextrct task of FTOOLS. Using runpcabackest command,the corresponding background light curves were generated fromstandard-2 data by using the bright background model providedby instrumentation team. Source and background spectra were ex-tracted from standard-2 data by using saextrct task for all theseobservations. The response matrices were created for each of theobservations by using pcarsp task. In addition to PCA, the HEXTEdata were also analysed to get hard X-ray spectrum from cluster-Bfor the observations carried out during 2006 giant outburst. Stan-dard procedures were followed to extract source and backgroundspectra and corresponding response matrices. Dead-time correctionwas also applied on the HEXTE spectra.

2.2 NuSTAR

NuSTAR is the first hard X-ray imaging observatory which waslaunched in 2012 June (Harrison et al. 2013). It consists of twoidentical grazing angle focusing telescopes FPMA and FPMB,operating in the range of 3–79 keV. A target of opportunity

observation (ID: 90201029002) was performed for the pulsar on2015 July 25 for an effective exposure of ∼57 ks. Though the ob-servation was carried out at an orbital phase where Type I X-rayoutburst was expected, there was no significant X-ray activity ob-served in the Swift/BAT monitoring light curve (Fuerst et al. 2016).Using NUSTARDAS software v1.4.1 of HEASOFT, we have reprocessedthe data and generated barycentric corrected light curves, spectraand response matrices and effective area files. The source prod-ucts were estimated from a circular region of 120 arcsec around thecentral source from FPMA and FPMB event data. The backgroundlight curve and spectra were also accumulated in a similar mannerby considering a circular region of 90 arcsec away from the source.

3 TIMING A NA LY SIS

3.1 Luminosity dependent pulse profiles

As described above, source and background light curves in 2–60 keV range were extracted at a time resolution of 0.125 s fromRXTE/PCA data. The barycentic correction was applied to the back-ground subtracted light curves by using faxbary task of FTOOLS. Theχ2-maximization technique was used to estimate the spin periodof the neutron star for all epochs of RXTE observations used inthe present analysis. We generated pulse profiles of the pulsar byfolding the light curves at respective spin periods. The epochs usedfor folding were chosen close to the beginning of the observationsand in such a manner that all the pulse profiles are aligned at theirminima for a better comparison. In order to understand the emissiongeometry, we were interested to explore the shape of pulse profilesover a wide range of source luminosity. The observations obtainedfrom various Type I outbursts and 2006 giant outburst are used inour study. Fig. 1 shows the 2–60 keV pulse profiles of the pulsarin increasing trend of luminosity, starting from ∼3.8 × 1036 to2.6 × 1038 erg s−1. The numbers quoted in each panel of Fig. 1 rep-resent the 3–30 keV source luminosity (in the unit of 1037 erg s−1;first number) and the beginning of the corresponding observation(in MJD; second number). The luminosity of the source duringeach of the observations was calculated based on the spectral fitting(see Section 4.1). It is clear from the figure that the pulse profilesare strongly dependent on source luminosity. At lower luminosity(∼1036 erg s−1), the pulse profile consisted of a significant peak atpulse phase of ∼0.3 along with a minor peak at ∼0.7 phase. Sharpdip-like features were also visible in the profiles below 0.2 phase.With increase in luminosity, the secondary minor peak starts evolv-ing into prominence and becomes comparable to the primary peak

MNRAS 472, 3455–3466 (2017)Downloaded from https://academic.oup.com/mnras/article-abstract/472/3/3455/4104638/Decade-long-RXTE-monitoring-observations-of-Be-Xby Physical Research Laboratory, Ahmedabad useron 17 October 2017

3458 P. Epili et al.

Figure 1. Pulse profiles of EXO 2030+375 obtained from Type I and Type II outbursts at different luminosities in 2–60 keV range. These profiles are generatedby folding the RXTE/PCA light curves at respective pulse period estimated separately for each of the observations. The epochs used for folding the light curveswith corresponding pulse period were considered close to the beginning of the observation and adjusted manually to align the pulse profiles for comparison. Thenumbers quoted in left and right side of each panel denote the 3–30 keV luminosity (in units of 1037 erg s−1) and beginning of the corresponding observation(in MJD) of the pulsar, respectively. Two pulses are shown in each panel for clarity. The error bars represent 1σ uncertainties.

(at ∼0.3 pulse phase) at a luminosity of ∼(4–7) × 1037 erg s−1. Be-yond this, the second peak remains significant whereas the strengthof first peak gradually decreases and finally disappears from thepulse profile at a luminosity of ≥1.6 × 1038 erg s−1. Multiple ab-sorption dips clearly appeared in the pulse profiles at certain pulsephases during the evolution with source luminosity. It is worth men-tioning that the shape of pulse profiles at extreme ends of observedsource luminosity (first and last panels of Fig. 1) is relatively similarwith a significant phase shift.

3.2 Pulse profiles during Type I and Type II outbursts

To investigate the changes in observed properties of the pulsarduring Type I and Type II X-ray outbursts, pulse profiles at variousepochs were generated and shown in Fig. 2. The data from severalType I outbursts were used at appropriate flux level to compare withthe observations at same intensity level, carried out during the 2006June giant outburst. Left-hand panels in the figure show the pulseprofiles obtained from several observations during the 2006 giantoutburst whereas the right-hand panels show the profiles obtainedfrom various Type I outbursts at comparable source luminosity.From the figure, it can be noted that (i) shape of pulse profiles aresimilar at comparable source intensity, irrespective of type of X-ray

outbursts; and (ii) evolution of pulse profiles with luminosity duringboth type of outbursts are also similar as remarked in Fig. 1. Thissignifies that pulse profiles of EXO 2030+375 are independent ofthe type of X-ray outbursts.

A thorough investigation of pulse profiles of the pulsar obtainedfrom observations performed during Type II outburst was also car-ried out. The motivation was to probe the evolution and possiblechange in the emission geometry at the same intensity during therising and declining part of the 2006 giant outburst. Due to largenumber of pointings in 2006 June, we were able to trace the pulsarbeam function at both phases of the outburst. For this, the pulse pro-files were generated at luminosity ranging from 1037 to 1038 erg s−1.Our results showed that the profiles during the rising and decliningphases of the outburst are similar at comparable luminosities. Thisbehaviour was analogous to as seen in Fig. 2.

3.3 Peculiar narrow and sharp absorption dip

We extracted a light curve with a time resolution of 0.1 s in 3–60 keVrange by using NuSTAR observation of the pulsar. As describedabove, standard data reduction procedure was followed to extractthe source and background light curves. From the background sub-tracted and barycentric corrected light curve, spin period of the

MNRAS 472, 3455–3466 (2017)Downloaded from https://academic.oup.com/mnras/article-abstract/472/3/3455/4104638/Decade-long-RXTE-monitoring-observations-of-Be-Xby Physical Research Laboratory, Ahmedabad useron 17 October 2017

RXTE observations of EXO 2030+375 3459

Figure 2. Pulse profiles of EXO 2030+375 during the Type II outburst in 2006 (left-hand panels) and during normal Type I outbursts (right-hand panels)in 2–60 keV range at comparable luminosities. The 3–30 keV luminosity of the pulsar (in 1037 erg s−1 units) and beginning of corresponding observation (inMJD) are quoted in left and right side of each panel. Two pulses are shown in each panel for clarity. The error bars represent 1σ uncertainties.

pulsar was estimated by using efsearch task of FTOOLS and foundto be 41.2932(2) s. Using this period, the pulse profile of the pul-sar was generated by folding the light curve at 128 phase bins andshown in the top panel of Fig. 3. Other panels of the figure show theprofiles from RXTE observations in the 2–60 keV range in increas-ing order of source luminosity. A narrow and sharp dip like structurein 0.1–0.2 phase range can be clearly seen in the profiles when thepulsar luminosity was ∼1036 erg s−1 (NuSTAR observation, also re-ported in Fuerst et al. 2017). This peculiar feature was also detectedin the pulse profile obtained from XMM–Newton observation of thepulsar during the 2014 May outburst (Ferrigno et al. 2016). Fig. 3shows that this peculiar dip is luminosity dependent and can beseen in the profiles up to the luminosity ≤4 × 1037 erg s−1. Beyondthis luminosity, the feature merges into a broader dip in the profile(bottom panel of Fig. 3). The evolution of the peculiar feature withenergy also showed strong variation that can be traced up to 30 keVby using data from NuSTAR observation.

4 SP E C T R A L A NA LY S I S

4.1 Phase-averaged spectroscopy

To probe spectral characteristics of the pulsar and correspondingchanges with luminosity, we carried out spectral studies by us-ing data from RXTE observations. Source and background spec-tra were extracted by following standard procedure as described

in Section 2.1 for all the RXTE observations. Using appropriatebackground spectra and response matrices, the source spectra fromPCA detector were fitted by using XSPEC package. Data in 3–30 keVrange were used in the spectral fitting. A systematic uncertaintyof 0.5 per cent was added to the data. While fitting the data, weexplored several continuum models that are used to describe theenergy spectrum of accretion powered X-ray pulsars. These mod-els are high energy cutoff power law, cutoff power law, negativeand positive exponentiation cutoff power law and with more phys-ical model such as CompTT. In our analysis, we used an absorbedpower-law model with a high-energy cutoff to describe the contin-uum spectrum of the pulsar. This model has been frequently usedto express the broad-band spectrum of EXO 2030+375 (Reig &Coe 1999; Wilson et al. 2008).

It has been recently found that the continuum of Be/X-ray binarypulsars is noticeably affected by the presence of additional mat-ter at certain pulse phases during outbursts that cannot be simplyexplained by a single absorber (Naik et al. 2013; Jaisawal, Naik &Epili 2016). A partial covering absorption component along with thecontinuum model is generally used to describe the pulsar spectrumin phase-averaged as well as phase-resolved spectroscopy. In ourfitting, a high energy cutoff power-law model was unable to fit theobserved spectrum, specifically during bright phases of outbursts.Addition of a partial covering component improved the fitting fur-ther and yielded an acceptable value of reduced-χ2 (∼1). The ironfluorescence emission line at ∼6.4 keV was also detected in the

MNRAS 472, 3455–3466 (2017)Downloaded from https://academic.oup.com/mnras/article-abstract/472/3/3455/4104638/Decade-long-RXTE-monitoring-observations-of-Be-Xby Physical Research Laboratory, Ahmedabad useron 17 October 2017

3460 P. Epili et al.

Figure 3. A peculiar narrow dip (in 0.1–0.2 phase range) can be seen inthe pulse profiles of EXO 2030+375. The profile in top panel representsdata from the NuSTAR observation whereas other panels show profiles ob-tained from the RXTE observations. The 3–30 keV luminosity (in unit of1037 erg s−1) and the beginning of corresponding observation (in MJD) arequoted in left and right sides of each panel, respectively. Two pulses areshown in each panel for clarity. The error bars represent 1σ uncertainties.

pulsar spectrum. A partially covering absorbed power law withhigh energy cutoff model is mathematically expressed as

N (E) = e−NH1σ (E)(K1 + K2e−NH2σ (E))f (E)

where

f (E) = E−� f orE < Ec

= E−�e−(

E−EcEf

)f orE > Ec

where f(E) represents the high energy cutoff power-law model with� as the power-law photon index, Ec and Ef are the cutoff and foldingenergies in keV, respectively. The normalization constants K1 andK2 are in the units of photon keV−1 cm−2 s−1. NH1 and NH2 arethe equivalent galactic hydrogen column density and the additionalcolumn density (in units of 1022 atoms cm−2), respectively. σ (E)is the photoelectric absorption cross-section. The energy spectra ofEXO 2030+375 along with the best-fitting model (top panel) andresiduals (bottom panel) for six epochs of RXTE observations areshown in Fig. 4. Observation IDs and 3–30 keV pulsar luminosityduring these epochs of observations are quoted in the figure. Thebest-fitted parameters obtained from spectral fitting of data obtainedat these epochs are given in Table 2.

During 2006 June outburst, detection of a cyclotron absorptionline like feature at ∼11 keV in the pulsar spectrum was reported(Wilson et al. 2008). We used the same observation in this study toinvestigate the cyclotron line feature in the pulsar spectrum. Duringthis outburst, as the pulsar luminosity was very high, data fromPCA and HEXTE detectors were used to get a broad-band spectralcoverage. The 3–100 keV broad-band spectrum, obtained from theRXTE observation on MJD 53962.5, was fitted with a high energycutoff power-law model yielding a poor fit with a reduced χ2 of

Figure 4. Phase-averaged energy spectra of EXO 2030+375 at different luminosity levels, obtained from six epochs of RXTE observations during Type Iand Type II X-ray outbursts. The spectra were fitted with partial covering high energy cutoff model along with an iron emission line at ∼6.4 keV. The sourcespectrum and best-fitting model are shown in the top panel, whereas the contribution of residuals to χ2 at each energy bin are shown in the bottom panel foreach epoch of RXTE observations. The observation IDs (in italics) and corresponding source luminosity are quoted in the figure.

MNRAS 472, 3455–3466 (2017)Downloaded from https://academic.oup.com/mnras/article-abstract/472/3/3455/4104638/Decade-long-RXTE-monitoring-observations-of-Be-Xby Physical Research Laboratory, Ahmedabad useron 17 October 2017

RXTE observations of EXO 2030+375 3461

Table 2. Best-fitting spectral parameters with 1σ errors obtained from RXTE/PCA observations of EXO 2030+375 at six different luminosities. The best-fittingmodel consists of a partial covering high energy cutoff power-law model with a Gaussian component.

Parameters Observation IDs70074-01-28-00 80071-01-01-06 91089-01-02-11 92067-01-03-11 91089-01-17-06 91089-01-12-04

NH1a 5.9 ± 0.8 6.2 ± 0.5 5.2 ± 0.3 2.8 ± 0.5 1.5 ± 0.4 3.41 ± 0.25

NH2b – – – 139.3 ± 17.5 166.5 ± 5.5 241.9 ± 4.0

Covering fraction – – – 0.25 ± 0.02 0.35 ± 0.01 0.45 ± 0.01Photon index (�) 1.72 ± 0.06 1.45 ± 0.06 1.34 ± 0.03 1.43 ± 0.05 1.61 ± 0.02 1.80 ± 0.03Ecut (keV) 8.2 ± 0.8 7.6 ± 0.4 7.7 ± 0.2 8.0 ± 0.3 8.2 ± 0.2 7.8 ± 0.2Efold (keV) 31.9 ± 5.6 23.3 ± 1.8 23.3 ± 0.8 21.2 ± 0.9 23.5 ± 0.5 25.7 ± 0.8

Emission linesFeKα line energy (keV) 6.32 ± 0.16 6.50 ± 0.11 6.33 ± 0.05 6.51 ± 0.08 6.43 ± 0.03 6.52 ± 0.04Width of Fe line (keV) 0.1 0.1 0.1 0.15 0.31 ± 0.07 0.1

Source fluxFluxc (3–10 keV) 0.17 ± 0.03 0.88 ± 0.09 1.8 ± 0.9 6.6 ± 0.7 13.8 ± 0.7 25.0 ± 1.5Fluxc (10–30 keV) 0.16 ± 0.02 0.94 ± 0.10 2.2 ± 0.1 7.1 ± 0.8 12.7 ± 0.6 19.4 ± 1.2Fluxc (3–30 keV) 0.33 ± 0.04 1.82 ± 0.19 4.0 ± 0.20 13.7 ± 1.4 26.5 ± 1.3 44.4 ± 2.6

Source luminosityLX

d (3–30 keV) 0.20 ± 0.03 1.10 ± 0.12 2.38 ± 0.14 8.23 ± 0.89 15.96 ± 0.90 26.77 ± 1.72

Reduced χ2(d.o.f.) 0.99 (24) 1.01 (34) 1.04 (46) 1.00 (50) 1.08 (53) 1.11 (43)

Notes. aEquivalent hydrogen column density (in 1022 atoms cm−2 unit); bAdditional hydrogen column density (in 1022 atoms cm−2 unit); cAbsorption correctedflux in unit of 10−9 ergs cm−2 s−1; dThe 3–30 keV X-ray luminosity in the units of 1037 ergs s−1 assuming a distance of 7.1 kpc to the source.

>8. A broad absorption-like feature at ∼10 keV was detected in thespectral residual (panel C of Fig. 5). Various combination of modelssuch as high energy cutoff power-law, CompTT, NPEX along withother components such as blackbody or a partial absorber wereused to test reliability of the reported line. Addition of a partialcovering component to the above continuum models resolves thebroad feature with a reduced-χ2 close to 1. Therefore, a high en-ergy cutoff model along with a partial covering component wasused as best-fitting model in our analysis. We generated Crab-ratioto check the presence of absorption like feature in the pulsar spec-trum. The Crab-ratio is obtained by normalizing the pulsar spectrumwith the feature-less power-law spectrum of Crab pulsar to removethe presence of any uncertainties related to calibration and model(see also Jaisawal, Naik & Paul 2013). This ratio showed a highlyabsorbed spectrum along with a 6.4 keV iron emission line below10 keV. We did not find any signature of absorption feature in the10–20 keV range in the Crab-ratio (panel B of Fig. 5). Spectralresiduals obtained from fitting the pulsar spectral with differentcontinuum models are shown in panels C, D, E and F of Fig. 5. Theabsence of any absorption like feature in 10–20 keV range can beclearly seen in above panels. To check the presence/absence of thisabsorption like feature, we fitted 3–79 keV broad-band spectrum ofEXO 2030+375, obtained from a NuSTAR observation during anextended period of low activity in 2015 with a high energy cutoffmodel. Any emission/absorption like features were not seen in thespectrum.

Spectral parameters such as power-law photon index, cutoff en-ergy, folding energy, additional column density (NH2), coveringfraction, hardness ratio (ratio between 10–30 keV flux and 3–10 keVflux) obtained from spectral fitting of all RXTE observations of EXO2030+375 are shown with corresponding 3–30 keV luminosity inFig. 6. All these parameters showed intriguing trends with luminos-ity which had not been explored earlier. In the figure, one can noticethat the values of power-law photon index are distributed in threedistinct regions such as negative, constant and positive correlations

with source luminosity which suggest a direct measure of spec-tral transition in EXO 2030+375. At lower luminosity (≤1037ergs−1), the pulsar spectrum was relatively soft. A negative correla-tion between the power-law photon index and luminosity can beclearly seen for this regime. The value of photon index was foundto be varying between 1.2 and 1.8 (first panel of Fig. 6). Whenthe luminosity was in the range of 2–4 × 1037erg s−1, the distri-bution of values of photon index did not show any dependence onsource luminosity. With increase in source luminosity, the photonindex showed a positive correlation. It is, therefore, clear that thepulsar changes its spectral behaviour with change in luminosity.Flux ratio (ratio between flux in 10–30 keV range and 3–10 keVrange) also showed smooth transition with increase in pulsar lu-minosity (left bottom panel of Fig. 6). As mentioned earlier, whilefitting spectra from high flux level of the outbursts, a partially ab-sorbed component was included in the fitting model. This was re-quired in spectral fitting when the pulsar luminosity was above 3 ×1037erg s−1. In our spectral fitting, the maximum value of additionalcolumn density (NH2) obtained was as high as 250 × 1022 cm−2,which is significantly larger than the value of interstellar absorptioncolumn density in the source direction. From our fitting, the valuesof additional column density and covering fraction were found to bestrongly luminosity dependent (middle panels of Fig. 6). The cutoffenergy and folding energy did not show any noticeable changeswith the pulsar luminosity (last panels of Fig. 6). The parametersobtained from NuSTAR observation are also included in the figure.

4.2 A physical model to describe the pulsarcontinuum spectrum

To explore the physical properties of accretion column, we havefitted the Becker and Wolff (BW) model with the phase aver-aged spectra of the pulsar. This model is proposed by Becker& Wolff (2007) to explain the emission from accretion powered

MNRAS 472, 3455–3466 (2017)Downloaded from https://academic.oup.com/mnras/article-abstract/472/3/3455/4104638/Decade-long-RXTE-monitoring-observations-of-Be-Xby Physical Research Laboratory, Ahmedabad useron 17 October 2017

3462 P. Epili et al.

Figure 5. The 3–100 keV energy spectra of EXO 2030+375 obtained fromPCA and HEXTE detectors of RXTE at the peak of 2006 giant outburst(MJD 53962.50). Source spectra along with the best-fitting model e.g. apartial covering high energy cutoff model along with a Gaussian for ironemission line (panel A) and corresponding spectral residual (panel G) areshown. Panels C, D, E and F indicate the spectral residuals obtained byfitting pulsar spectra with a (i) high energy cutoff model, (ii) high energycutoff model with a blackbody, (iii) CompTT with a blackbody and (iv) apartial covering CompTT model with blackbody component, respectively,along with interstellar absorption component and a Gaussian function foriron emission line at 6.4 keV. Any signature of cyclotron absorption line atpreviously reported value of ∼11 keV is not seen in the spectral residuals.The Crab-ratio (panel B) also did not show any such feature in 10–20 keVrange.

X-ray pulsars by considering the effects of thermal and bulk Comp-tonizations in accretion column. It has been successful in explain-ing the broad-band spectra of bright X-ray pulsars such as 4U0115+63 (Ferrigno et al. 2009), 4U 1626-67 (D’Aı et al. 2017),Her X-1 (Wolff et al. 2016). According to this model, seed photonsoriginated due to bremsstrahlung, blackbody and cyclotron emis-sions undergo thermal and bulk Comptonization in the accretioncolumn. Comptonization of these seed photons with highly ener-getic electrons lead to the power law like resultant spectra with highenergy exponential cutoff.

Using this model, we have described the 3–70 keV broad-bandphase-averaged spectra of EXO 2030+375 at 23 different luminos-ity epochs, covering the range of 1036–1038 erg s−1. For a canonicalneutron star mass and radius, the BW model has six free param-eters, i.e. the diffusion parameter ξ , the ratio of bulk to thermalComptonization δ, the column radius r0, mass accretion rate M ,electron temperature Te and the magnetic field strength B. Amongthese parameters, the mass accretion rate M was estimated by usingthe observed source flux obtained from high energy cutoff empir-ical model and considering a source distance of 7.1 kpc (Wilson

et al. 2002). Since the column radius strongly depends on accretionrate (see equation 112 of Becker & Wolff 2007), the parameter M

was fixed at a given value while fitting the spectra. After fitting,the column radius was also fixed for getting better constraints onother spectral parameters. This was done carefully by analysing the2D contour plots between r0 & M , as similar to Ferrigno et al.(2009). The other BW model components such as normalizationsof bremsstrahlung, cyclotron and blackbody seed photons were alsokept fixed as suggested in the BW_cookbook.1 A partial coveringcomponent as required in empirical models was also needed to ex-plain the absorbed spectra of the pulsars during bright outbursts.The values of additional column density and covering fraction ob-tained from BW model were found to be consistent with the valuesobtained with the high energy cutoff power-law model. Therefore,we have not discussed these parameters in this section. An ironfluorescence line at ∼6.4 keV was also added in the continuum.Spectral parameters obtained after best fitting the pulsar spectrawith BW model are given in Table 3. The values of reduced χ2

obtained from our fitting, as given in Table 3, showed that the BWcontinuum model fits the data well in a wide luminosity range.For three different values of pulsar luminosity, broad-band energyspectra of the pulsar from PCA and HEXTE detectors, along withbest-fitted BW continuum model and Gaussian function for ironemission line are presented in top panels of the Fig. 7. The bottompanels in this figure show corresponding spectral residuals of thebest-fitted model. It can be seen that the residuals obtained fromfitting the pulsar spectra with BW model did not show any evidenceof presence of absorption like feature. This finding also supports thenon-detection of cyclotron line in EXO 2030+375, as discussed inthe above section of the paper.

Luminosity dependent variations in the parameters obtained afterfitting the pulsar spectra with the BW model are shown in Fig. 8. Aninteresting trend of parameter δ with luminosity was noticed in thethird panel of the figure. This parameter signifies the ratio of bulkto thermal Comptonization occurring in the accretion column. Thevalue of δ was found close to unity at luminosity ≤3–4 × 1037 ergss−1. This indicates that the effects of thermal and bulk Comptoniza-tion are nearly same in accretion column at lower luminosity ofthe pulsar. However, as the luminosity increases, bulk Comptoniza-tion starts playing a major role to column emissions and dominatesover by a factor of 20 times as observed at lower luminosity. Thecolumn radii r0 was found to be strongly dependent on luminosityor mass accretion rate. Moreover, the diffusion parameter ξ wasalso observed to vary with source intensity, showing a minimumvalue at higher luminosity. In addition to these, the electron plasmatemperature was changing in the range of 3 to 7 keV. The tempera-ture showed a gradual increase up to luminosity 4 × 1037 ergs s−1.Beyond this, a cooling of plasma temperature was observed. Thismay occur in the presence of strong radiation dominated accretionshock at which the infalling matter mostly bulk Comptonize theseed photons that carries plasma energy by diffusing through theside walls of accretion column. It leads to the settling of plasma inaccretion column at lower temperature. This model also providesan opportunity to constrain the magnetic field of the pulsar in therange of ∼4–6 × 1012 G (see Fig. 8). In some cases, magnetic fieldestimated from this model was found insensitive to upper value,though their lower estimate was easily constrained in all these ob-servations. Therefore, only best-fitted values without error bars arequoted in Table 3.

1 http://www.isdc.unige.ch/ferrigno/images/Documents/BW_distribution/BW_cookbook.html

MNRAS 472, 3455–3466 (2017)Downloaded from https://academic.oup.com/mnras/article-abstract/472/3/3455/4104638/Decade-long-RXTE-monitoring-observations-of-Be-Xby Physical Research Laboratory, Ahmedabad useron 17 October 2017

RXTE observations of EXO 2030+375 3463

Figure 6. Spectral parameters such as photon index (top left panel), additional column density (top middle panel), cutoff energy (top right panel), coveringfraction (bottom middle panel) and folding energy (bottom right panel) obtained from the spectral fitting of RXTE observations of EXO 2030+375 with apartial covering high energy cutoff power-law model during Type I and Type II outbursts, are shown with the 3–30 keV luminosity. Hardness ratio (ratiobetween 10–30 keV flux and 3–10 keV flux) with luminosity is also shown in bottom left panel of the figure. The parameters from NuSTAR observation aremarked with empty triangles. The error bars quoted for 1σ uncertainties.

Table 3. Best-fitting spectral parameters with 1σ errors obtained from RXTE/PCA and RXTE/HEXTE observations of EXO 2030+375 with BW model.

BW model parametersObs-Ids Luminositya M ξ δ B Te r0 Reduced χ2

(1037ergs s−1) (1017 g s−1) (1012) G (keV) (m) (d.o.f.)

91089-01-12-04b 28.04 ± 0.84 14.96 1.88 ± 0.03 14.38 ± 1.01 4.25 ± 0.20 3.95 ± 0.11 177.0 1.22(103)91089-01-12-05 28.22 ± 0.38 14.82 1.77 ± 0.02 17.57 ± 1.20 4.23 3.42 ± 0.10 165.0 1.04(80)91089-01-15-00 27.12 ± 0.53 14.59 1.82 ± 0.02 16.82 ± 1.19 4.58 3.83 ± 0.10 158.0 1.05(104)91089-01-14-03 26.36 ± 0.49 13.88 1.86 ± 0.03 17.03 ± 2.50 4.34 ± 0.25 4.01 ± 0.16 141.0 1.02(85)91089-01-09-03 19.39 ± 0.39 10.44 1.98 ± 0.03 11.24 ± 0.91 4.66 ± 0.22 4.53 ± 0.12 138.0 1.09(105)91089-01-09-00 17.11 ± 0.65 9.21 2.06 ± 0.05 9.64 ± 2.75 5.29 5.15 ± 0.40 116.0 1.09(104)91089-01-08-01 13.40 ± 0.48 7.22 2.13 ± 0.05 10.48 ± 1.10 5.59 5.61 ± 0.14 105.0 1.08(107)91089-01-07-00 11.63 ± 0.58 6.26 2.43 ± 0.17 4.22 ± 0.72 4.85 ± 0.35 5.56 ± 0.31 98.5 1.13(98)92067-01-03-13 10.09 ± 0.54 5.43 2.44 ± 0.16 4.17 ± 0.54 5.18 ± 0.36 5.97 ± 0.40 74.0 1.09(90)92067-01-03-11 8.17 ± 0.98 4.40 2.36 ± 0.29 4.30 ± 1.16 5.33 ± 0.35 6.14 ± 0.43 35.6 1.16(95)92067-01-03-00 6.86 ± 1.47 3.69 3.11 ± 0.73 2.09 ± 1.07 5.03 6.32 ± 0.16 28.2 1.10(99)93098-01-03-05 5.72 ± 0.70 3.25 1.91 ± 0.16 7.90 ± 1.38 4.85 4.85 ± 0.41 26.0 1.08(87)93098-01-01-01 4.86 ± 0.46 2.79 2.08 ± 0.15 5.50 ± 1.10 4.35 ± 0.21 4.88 ± 0.27 25.5 1.05(111)92422-01-25-06b 4.74 ± 0.74 2.66 3.01 ± 1.19 2.43 ± 1.20 5.39 ± 0.45 6.68 ± 0.36 24.5 1.11(101)93098-01-01-00 3.90 ± 0.83 2.34 5.08 ± 1.24 0.97 ± 0.35 4.78 6.32 ± 0.07 19.6 1.08(102)92422-01-05-04 3.82 ± 0.84 2.22 7.37 ± 3.18 0.64 ± 0.21 4.82 6.53 ± 0.05 19.2 0.99(96)93098-01-02-04 3.71 ± 1.08 1.98 7.26 ± 2.86 0.62 ± 0.27 4.70 6.34 ± 0.07 23.4 1.03(89)92422-01-14-03 3.07 ± 0.75 1.74 6.11 ± 1.95 0.77 ± 0.29 4.50 6.09 ± 0.07 17.5 1.07(90)80071-01-01-11 1.41 ± 0.01 0.81 6.49 ± 1.05 0.66 ± 0.13 3.74 5.24 ± 0.02 17.5 1.06(87)80071-01-01-06 1.20 ± 0.01 0.67 8.15 ± 2.50 0.52 ± 0.20 3.77 5.29 ± 0.03 16.0 1.08(86)80071-01-01-00 1.14 ± 0.01 0.64 7.11 ± 1.65 0.60 ± 0.17 4.02 5.69 ± 0.03 18.4 1.19(97)80071-01-01-01b 0.88 ± 0.01 0.48 7.89 ± 1.80 0.57 ± 0.15 3.75 5.32 ± 0.02 17.4 1.12(87)80071-01-01-020 0.92 ± 0.01 0.44 10.29 ± 2.58 0.39 ± 0.16 3.69 5.28 ± 0.01 15.0 1.03(72)

Notes. aThe 3–70 keV luminosity in the units of 1037 ergs s−1 by assuming a distance of 7.1 kpc.bIndicates the observation Ids from which extracted spectra are shown in Fig. 7.

5 D ISC U SSION

Be/X-ray binaries show two types of X-ray outbursts such as TypeI and Type II outbursts. These outbursts are known to be due tocapture of huge amount of matter by the neutron star from the cir-cumstellar disc of the companion Be star at periastron passage (TypeI) or due to evacuation/truncation of Be circumstellar disc (Type II).

During these episodes, X-ray emission from the neutron star getstransiently enhanced by a factor of ten or more. Normal (Type I)outbursts are found to be less luminous (1036 − 37 erg s−1) whereasduring giant (Type II) outbursts, the luminosity reaches close toor above the Eddington luminosity of a neutron star (Negueruelaet al. 1998). EXO 2030+375 is a unique Be/X-ray binary system

MNRAS 472, 3455–3466 (2017)Downloaded from https://academic.oup.com/mnras/article-abstract/472/3/3455/4104638/Decade-long-RXTE-monitoring-observations-of-Be-Xby Physical Research Laboratory, Ahmedabad useron 17 October 2017

3464 P. Epili et al.

Figure 7. Phase-averaged energy spectra of EXO 2030+375 at three distinct luminosities obtained during Type-I and Type-II X-ray outbursts obtained fromPCA and HEXTE detectors. The spectra of pulsar was fitted with BW model (Ferrigno et al. 2009) along with an iron line at ∼6.4 keV and partial coveringcomponent (top panel of figure). Corresponding spectral residuals are shown in bottom panels of the figure. Any absorption like feature was not seen in10–20 keV energy range of the pulsar spectra.

Figure 8. Spectral parameters obtained from the fitting of phase averagedspectra of EXO 2030+375 with BW model at different luminosities. Thetop, second and third panels of the figure show the mass accretion rate,diffuse rate and the ratio of bulk to thermal Comptonization in accretioncolumn, respectively. While the fourth, fifth and sixth panels indicate the lu-minosity variation of magnetic field, plasma temperature and column radius,respectively.

that displays Type I outbursts almost at each periastron passage.However, the pulsar showed Type II outbursts only twice (in 1985and 2006) since its discovery. Type I outbursts are short lived, cov-ering about 20–30 per cent of orbital period in contrast to giantoutbursts which lasted for more than three months in both the occa-sions. The peak luminosity during normal outburst varies dependingon the evolutionary state of the Be circumstellar disc. It is believedthat the neutron star accretes matter from the circumstellar disc atthe periastron passage and gives rise to Type I X-ray outbursts.As the Type II X-ray outbursts are very rare, the origin for TypeII outburst remains unclear. It is thought that dramatic expansionof the circumstellar disc around the Be star or instabilities in thecircumstellar disc leads to such major events (Okazaki & Negueru-ela 2001; Reig 2011). Several analytical and numerical studies havebeen performed to understand the behaviour of transient accretionby considering a misalignment between the orbital plane and theeccentric warped Be circumstellar disc (Okazaki, Hayasaki & Mori-tani 2013; Martin et al. 2014). These studies showed that accretiontime-scales longer than the orbital period and higher luminosity asobserved during Type II outbursts are possible in such scenario.Using large number of RXTE observations of EXO 2030+375 dur-ing Type I and Type II outbursts, we have explored the evolutionof pulse profile and its characteristics as a function of luminosityas well as type and phase of outbursts. The changes in spectralproperties and its transition are also studied in detail in the paper.

5.1 Pulse profiles

Pulse profiles of Be/X-ray binary pulsars are generally complex dueto the presence of multiple absorption dips at various spin phasesof the pulsar. We have explored the evolution of pulse profiles ofEXO 2030+375 over a wide range of luminosity, starting from∼1036 to 1038 erg s−1 by using large number of RXTE observa-tions of the pulsar. The profiles are found to be strongly luminositydependent which has also been reported from previous studies (Par-mar et al. 1989b; Klochkov et al. 2008; Naik et al. 2013; Naik &Jaisawal 2015). One of the interesting results we obtained is that theprofiles are similar in shape at identical luminosities irrespective oftypes and phases of outbursts. This indicates that the beam pattern ofthe pulsar does not depend on outburst characteristics, although dis-tinct mechanism governs the origin of Type I and Type II outbursts.Accretion rate or luminosity is the vital factor that decide the shapeof pulse profile of the pulsar. At lower luminosity (≤1037 erg s−1;

MNRAS 472, 3455–3466 (2017)Downloaded from https://academic.oup.com/mnras/article-abstract/472/3/3455/4104638/Decade-long-RXTE-monitoring-observations-of-Be-Xby Physical Research Laboratory, Ahmedabad useron 17 October 2017

RXTE observations of EXO 2030+375 3465

sub-critical regime), the X-ray emission is believed to be originatedfrom the hotspot along the accretion column in the form of pencilbeam (Sasaki et al. 2010 and references therein). This beam ge-ometry leads to the formation of a single peaked profile as seen inour study at lower luminosity. In presence of a radiation dominatedshock at higher luminosity, the profile changes from single to doublepeaked. A mixture of pencil and fan beam patterns can contribute tothe double peak structure of pulse profiles which is clearly seen inEXO 2030+375 in the luminosity range of ∼(3–12) × 1037 erg s−1.During the giant outbursts, mass accretion rate increases beyondthe critical luminosity that shifts the height of radiative shock in theaccretion column. As a result, accreted matter is obstructed by dom-inating radiation pressure above the shock (Becker & Wolff 2007).The photons beyond this point mostly diffuses through the side wallof the accretion column, forming a fan-beam geometry (Klochkovet al. 2008). Therefore, at higher luminosity (>1.2 × 1038 erg s−1),the pulse profiles of the pulsar could be purely due to the fan beampattern.

The energy dependent pulse profiles in EXO 2030+375 has alsobeen reported earlier (Reig & Coe 1999; Klochkov et al. 2008; Naiket al. 2013; Naik & Jaisawal 2015). The shape of the pulse profiles ofthe pulsar is complex due to the presence of multiple peaks/dips insoft X-rays. A single peak profile is generally seen at hard X-rays.Sasaki et al. (2010) modelled the energy resolved pulse profilesduring a giant outburst to understand the emission componentsfrom the magnetic poles of the pulsar. An asymmetric profile wasexplained by considering a moderate distortion in the magnetic field,which can locate one of the accretion column at relatively closeto the line of sight. The composite emissions from both the polesresulted an asymmetry in the profile due to deformity in the locationof columns. Suzaku observations of the pulsar during 2007 and 2012Type I outbursts showed nearly symmetrical profiles along withmultiple absorption dips at certain pulse phases (Naik et al. 2013;Naik & Jaisawal 2015). These dips showed strong dependence onenergy and luminosity. Presence of additional matter in the formof narrow streams that are phase locked to the pulsar is believedto be the cause of these features in pulse profiles. Such absorptionlike features are also seen in pulse profiles of other Be/X-ray binarypulsars such as A 0535+262 (Naik et al. 2008), GRO J1008-57(Naik et al. 2011) and GX 304-1 (Jaisawal et al. 2016) duringoutbursts. During 2007 outburst of EXO 2030+375, these dipswere present in the profiles up to ∼70 keV (Naik et al. 2013). Phaseresolved spectroscopy confirmed the presence of additional densematter at same phases of the profile. In addition to absorption dips,sometimes a narrow and sharp dip-like feature was detected in thepulse profiles at low luminosity (Ferrigno et al. 2016). This featurewas interpreted as due to self-absorption from the accretion column.This peculiar feature was also detected in the pulse profiles of EXO2030+375 and found to be luminosity dependent. The feature waspresent in the pulse profiles when the pulsar luminosity was below4 × 1037 erg s−1 (critical luminosity regime). It is probable that atlower luminosity, the pencil beam propagating across the magneticaxis interacts with the accretion column directly and produces adip-like structure in the pulse profile. A significant contributionfrom fan beam may change the emission geometry and lead to theabsence of this feature beyond the critical luminosity, as seen in ourstudy.

5.2 Spectroscopy

The broad-band energy spectrum of accretion powered X-ray pul-sar is known to be originated due to the inverse Comptonization

of soft X-ray photons emitted from the hot spots on the surface(Becker & Wolff 2007). The continuum is generally described withsimple models such as high energy cutoff power law, NPEX, expo-nential cutoff power law etc. despite of the complex phenomenonoccurring in the accretion column. A physical model, known as BWmodel, was also used in our study to understand the spectral prop-erties of the pulsar at different luminosity levels. We carried outspectral analysis of a large number of RXTE pointed observationsof Be/X-ray binary pulsar EXO 2030+375, spanned over a decadeby using a high energy cutoff power-law model along with a par-tial absorbing component and a Gaussian function for the 6.4 keViron emission line. The RXTE observations of the pulsar providedopportunity to trace spectral evolution of the pulsar at various lumi-nosity levels during Type I and Type II X-ray outbursts from 1996to 2011. Parameters obtained from the spectral fitting showed veryinteresting variation with luminosity. The photon index was foundto exhibit three distinct patterns with luminosity indicating signa-tures of spectral transition between sub-critical and super criticalstates. All three regimes are reflected in the pattern of the pulseprofiles and are interpreted as due to different beam pattern at threedifferent luminosity ranges.

A negative correlation was seen between power-law photon in-dex and pulsar luminosity in sub-critical regime (below the criticalluminosity) where spectrum was relatively hard. In this condition,the broad-band X-ray emission is considered to have originatedfrom a hot mount on the neutron star surface (Basko & Sun-yaev 1976; Becker et al. 2012). Critical luminosity is associatedwith the transition between two accretion scenarios. In our study,we detected a plateau like region in the distribution of power-lawphoton index in luminosity range of ∼(2–4) × 1037 erg s−1. Thisluminosity range can be considered as the critical luminosity forEXO 2030+375. The critical luminosity regime has been exploredfor other accretion powered X-ray pulsars such as 4U 0115+63,V 0332+53, Her X-1, A 0535+26 and GX 304-1 in luminosityrange of ∼(2–8) × 1037 erg s−1 (Becker et al. 2012 and referencetherein). A positive correlation between photon index and luminos-ity was detected above the critical luminosity. This occurs becauseof the dominating role of shock in the accretion column which ef-fectively reduces the velocity of energetic electrons. In this case,pulsar spectrum appears soft due to lack of bulk Comptonizationof photons with accreting electrons (Becker et al. 2012). A posi-tive correlation between photon index and luminosity therefore isobserved in supercritical regime. During 1985 giant outburst, thephoton index was also proportional to luminosity, indicating that thepulsar was accreting above the critical limit (Reynolds, Parmar &White 1993).

The hardness ratio (ratio between 10–30 keV flux and 3–10 keVflux) also showed similar kind of transition with luminosity. Thisshowed increasing trend till the pulsar luminosity reaches its criticalvalue beyond which a decreasing trend was observed. It showed thatthe pulsar emission was relatively hard in the sub-critical luminosityregion. A softening in the spectrum was observed above the criticalluminosity as discussed in the above section. The folding energy wasfound to vary with luminosity. This spectral parameter represents theplasma temperature in the emission region. At lower luminosity, thevalue of folding energy was relatively constant although a few highvalues (with large error bars) were also evident. The high values insub-critical luminosity regime may correspond to the deep regionsof accretion column. The value of folding energy was increasingbeyond the critical luminosity. It is expected that increasing massaccretion in supercritical region can produce the high temperatureplasma in the presence of shock.

MNRAS 472, 3455–3466 (2017)Downloaded from https://academic.oup.com/mnras/article-abstract/472/3/3455/4104638/Decade-long-RXTE-monitoring-observations-of-Be-Xby Physical Research Laboratory, Ahmedabad useron 17 October 2017

3466 P. Epili et al.

The magnetic field of the pulsar can be investigated by usingobserved cyclotron resonance scattering features in the broad-bandspectrum. Cyclotron resonance scattering features appear due tothe resonant scattering of electrons with photons in the presence ofmagnetic field (Caballero & Wilms 2012). These absorption likefeatures appear in the hard X-ray spectrum of accretion powered X-ray pulsars with magnetic field in the order of 1012 G. The detectionof these features allow us to directly estimate the magnetic field ofpulsar (Jaisawal & Naik 2017). Detection of a cyclotron absorptionline at ∼11 keV in the pulsar spectrum obtained from RXTE ob-servation was reported earlier (Wilson et al. 2008). Using the samedata set, we attempted to explore the cyclotron line feature further.However, our results showed that this feature was model dependent,only seen in a single cutoff based model. This discards the detec-tion of cyclotron scattering feature in EXO 2030+375. The featurewas also not detected in the pulsar spectra obtained from NuSTARobservations, though it was carried out during low intensity phase.Moreover, our studies based on the BW model showed a constrainon pulsar magnetic field which is in the range of ∼4–6 × 1012 G.For such strength of magnetic field, a cyclotron feature is expectedto be observed in the 40–60 keV energy range of pulsar spectrum.We have not observed any such features in the above energy rangeusing RXTE observations, although a claim of cyclotron line at ∼36or 63 keV have been made earlier in the pulsar (Reig & Coe 1999;Klochkov et al. 2008). With high capability of new generation satel-lites such as Astrosat and NuSTAR, the cyclotron line feature can beinvestigated during an intense X-ray outburst.

6 SU M M A RY A N D C O N C L U S I O N

We have carried out a detailed timing and spectral analysis oftransient Be/X-ray binary pulsar EXO 2030+375 during severalType I and Type II outbursts during 1996–2011 with RXTE. Anabsorbed power law modified with a high-energy cutoff along witha partial absorber and a Gaussian component was used to explainthe 3–30 keV energy spectrum of the pulsar. The pulse profileswere found to be strongly luminosity dependent at wide range of∼1036−38 erg s−1. Observed changes in the shape of pulse profilesare attributed to the change in emission geometry of pulsar. Pulsaremission was dominated by pencil beam at lower luminosity whichswitched to fan beam at higher luminosity. The different shape ofpulse profiles at both the extremes are interpreted as due to differentbeam patterns. However, a mixture of both patterns are seen in thecritical luminosity regime. The profiles were observed to be inde-pendent of Type I and Type II outbursts and their phases. Spectralparameters also showed the signatures of emission transition. Basedon luminosity, a transition from sub-critical to supercritical regimeis seen in the photon index. These changes are explained in termsof changes in the emission geometry across the critical luminosity.At the brighter phases the presence of additional matter nearby thepulsar is observed due to the effect of higher accretion rate. Basedon the physical modelling of the continuum spectrum, the magneticfield of pulsar can be estimated to be ∼4–6 × 1012 G.

AC K N OW L E D G E M E N T S

We thank the referee for his/her constructive suggestions that im-proved the paper. The research work at Physical Research Labora-tory is funded by the Department of Space, Government of India.This research has made use of data obtained through HEASARCOnline Service, provided by NASA/GSFC, in support of NASAHigh Energy Astrophysics Programs.

R E F E R E N C E S

Basko M. M., Sunyaev R. A., 1976, MNRAS, 175, 395Becker P. A., Wolff M. T., 2007, ApJ, 654, 435Becker P. A. et al., 2012, A&A, 544, 123Caballero I., Wilms J., 2012, Mem. Soc. Astron. Ital., 83, 230Coe M. J., Payne B. J., Longmore A., Hanson C. G., 1988, MNRAS, 232,

865D’Aı A., Cusumano G., Del Santo M., La Parola V., Segreto A., 2017,

MNRAS, 470, 2457Ferrigno C., Becker P. A., Segreto A., Mineo T., Santangelo A., 2009, A&A,

498, 825Ferrigno C., Pjanka P., Bozzo E., Klochkov D., Ducci L., Zdziarski A. A.,

2016, A&A, 593, A105Fuerst F., Wilson-Hodge C. A., Kretschmar P., Kajava J., Kuehnel M., 2016,

Astron. Telegram, 8835Fuerst F. et al., 2017, A&A, preprint (arXiv:1707.06496)Harrison F. A. et al., 2013, ApJ, 770, 103Jahoda K., Swank J. H., Giles A. B., Stark M. J., Strohmayer T., Zhang W.,

Morgan E. H., 1996, in Siegmund O. H., Gummin M. A., Proc. SPIEConf. Ser. Vol. 2808, EUV, X-Ray, and Gamma-Ray Instrumentation forAstronomy VII. SPIE, Bellingham, p. 59

Jahoda K., Markwardt C. B., Radeva Y., Rots A. H., Stark M. J., Swank J.H., Strohmayer T. E., Zhang W., 2006, ApJS, 163, 401

Jaisawal G. K., Naik S., Paul B., 2013, ApJ, 779, 54Jaisawal G. K., Naik S., Epili P., 2016, MNRAS, 457, 2749Jaisawal G. K., Naik S., 2017, in Serino M., Shidatsu M., Iwakiri W.,

Mihara T., eds, 7 years of MAXI: monitoring X-ray transients, 2016, p.153 (arXiv:1705.05536)

Klochkov D., Santangelo A., Staubert R., Ferrigno C., 2008, A&A, 491,833

Kretschmar P. et al., 2016, Astron. Telegram, 9485Krimm H., Barthelmy S., Gehrels N., Markwardt C., Palmer D., Sanwal D.,

Tueller J., 2006, Astron. Telegram, 861Levine A. M., Bradt H., Cui W., Jernigan J. G., Morgan E. H., Remillard R.,

Shirey R. E., Smith D. A., 1996, ApJ, 469, L33Martin R. G., Nixon C., Armitage P. J., Lubow S. H., Price D. J., 2014, ApJ,

790, L34Motch C., Janot-Pacheco E., 1987, A&A, 182, L55Naik S., Jaisawal G. K., 2015, Res. Astron. Astrophys., 15, 537Naik S. et al., 2008, ApJ, 672, 516Naik S., Paul B., Kachhara C., Vadawale S. V., 2011, MNRAS, 413, 241Naik S., Maitra C., Jaisawal G. K., Paul B., 2013, ApJ, 764, 158Negueruela I., Reig P., Coe M. J., Fabregat J., 1998, A&A, 336, 251Okazaki A. T., Negueruela I., 2001, A&A, 377, 161Okazaki A. T., Hayasaki K., Moritani Y., 2013, PASJ, 65, 41Parmar A. N., White N. E., Stella L., Izzo C., Ferri P., 1989a, ApJ, 338,

359Parmar A. N., White N. E., Stella L., 1989b, ApJ, 338, 373Paul B., Naik S., 2011, Bull. Astron. Soc. India, 39, 429Reig P., 2011, Ap&SS, 332, 1Reig P., Coe M. J., 1999, MNRAS, 302, 700Reynolds A. P., Parmar A. N., White N. E., 1993, ApJ, 414, 302Rothschild R. E. et al., 1998, ApJ, 496, 538Sasaki M., Klochkov D., Kraus U., Caballero I., Santangelo A., 2010, A&A,

517, A8Stella L., White N. E., Rosner R., 1986, ApJ, 308, 669Stollberg M. T., 1997, PhD thesis, Univ. AlabamaSun X.-J., Li T.-P., Wu M., Cheng L.-X., 1994, A&A, 289, 127Wilson C. A., Finger M. H., Coe M. J., Laycock S., Fabregat J., 2002, ApJ,

570, 287Wilson C. A., Fabregatet J., Coburn W., 2005, ApJ, 620, L99Wilson C. A., Finger M. H., Camero-Arranz A., 2008, ApJ, 678, 1263Wolff M. T. et al., 2016, ApJ, 831, 194

This paper has been typeset from a TEX/LATEX file prepared by the author.

MNRAS 472, 3455–3466 (2017)Downloaded from https://academic.oup.com/mnras/article-abstract/472/3/3455/4104638/Decade-long-RXTE-monitoring-observations-of-Be-Xby Physical Research Laboratory, Ahmedabad useron 17 October 2017