1. Introduction
A high energy phenomena, the so called supernova explosion, takes place at the end of stellar evolution cycle of massive stars [
49]. A huge amount of energy, released by the gravitationally collapsing stellar core, produces shocks that eject the outer layers of the progenitor star. This leads to the formation of relativistic compact objects such as a white dwarf, a neutron star or a black hole with a very high matter density [
2]. The masses of white dwarfs and neutron stars are in the range 0.18-1.3M
and 1-3M
respectively. The white dwarfs are approximately Earth-sized while neutron stars have radius around 10km [
3,
4]. Such objects reach extreme physical conditions of density, temperature, pressure, strong surface magnetic fields and fast rotations. Generally, neutron stars are formed after the core collapse of stars with initial masses in the range 8-20M
and the corresponding supernova explosion asymmetrically ejects the matter at a very high velocity of ∼ 10
m
[
5,
6]. Asymmetric ejection of matter imparts natal kicks to the neutron star due to conservation of momentum. Consequently, the resulting kick force causes the rotation of neutron star unless it is exerted exactly head-on at the center of star. However, the exact physical process which kicks the neutron star initially at birth has not been clearly understood and is still under debate [
7,
8,
9]. Rapidly spinning/rotating neutron stars with a surface magentic field strength above ∼ 10
G are referred to as
pulsars [
10,
11]. The strength and structure of magnetic field play a significant role in describing the observational properties and ages of pulsars. Numerical simulations suggest that isolated pulsars (spherical neutron star with dipole magnetic field) follow the
spin-down law of the form [
12]
where
P is spin period of the pulsar,
is time-derivative of
P or spin-down rate,
and
I are respectively the magnetic dipole moment and moment of inertia of the neutron star,
is the angle between magnetic and rotation/spin axes and
c is speed of light in vacuum. Angle
is also referred to as the
magnetic angle or the
magnetic obliquity of the neutron star. The values of
close to 0
and 90
correspond to the aligned and orthogonal pulsars respectively. The field lines of strong magnetic field of pulsar are anchored to the outer crust of the neutron star. Resulting which, the rotation axis and frequency of rotation of the surface magnetic field are same as the pulsar rotation axis and its spin frequency respectively. A strong electric potential is produced along the magnetic poles due to the co-rotation of the magnetic field lines near the surface and outer crust of the neutron star. This electric potential generates a powerful current of charged particles traveling along the field lines in the magnetosphere [
13,
14]. The charged particles moving along the magnetic field lines create a radiation cone which is centered on the magnetic axis. Since, the magnetic axis does not coincide with the spin/rotation axis, series of periodic pulses are detected due to rotation of neutron star if the radiation cone sweeps past our line of sight. Therefore, the pulsars are considered as astronomical lighthouse. The pulsed radiation measured from pulsars spans over the entire electromagnetic spectrum rangeing from radio to high energy gamma-rays and is very precisely characterized by their spin period (
P).
Historically, the first pulsar
(CP:Cambridge Pulsar) or
with
1.34s was discovered by Jocelyn Bell and Antony Hewish in 1967 at the Cambridge University using an array of diploes tuned to 81.5 MHz radio frequency. However, the official announcement was made in the February 24, 1968, edition of the prestigious
Nature journal [
10] and the Nobel Prize in Physics was awarded to Antony Hewish in 1974 for the discovery of pulsars. This breakthrough result was interpreted as the radial pulsations of either a neutron star or a white dwarf. Subsequently, about 30 pulsars were detetced using radio observations in a span of two years and a sudden decrease in the pular spin period (pulsar glitch) by about 3 parts in a million was also discovered in 1969 [
15,
16]. The discovery of first binary pulsar PSR B1913+16 by Russell Hulse and Joe Taylor in 1974 provided the first accurate measurement of neutron star mass [
17]. The next important discovery of a millisecond pulsar PSR B1937+214 with
1.558ms resolved the puzzle of extremely steep radio spectrum of pulsars in 1982 [
18]. A unique binary radio pulsar PSR J0737–3039A/B with
22ms and an orbital period of 2.4 hours was discovered in 2003 [
19]. This binary system comprises two neutron stars of different ages and their orbit can be viewed exactly edge-on. Discovery of this eclipsing binary pulsar system is hailed as a milestone both in observational search as well as exploring the fundamental physics. Over the past 50 years, the pulsar astronomy has made a very impressive progress as the number of known pulsars in our Milky-Way Galaxy has rapidly grown to more than 3000 including isolated and binary pulsars with diverse characteristics over the entire electromagentic spectrum
1. The distribution of latitudes and longitudes of the known pulsar population in the Milky-Way Galaxy are shown in
Figure 1. It is evident that the Galactic latitude of most of the pulsars is close to zero. This suggests that the Galactic plane is populated with the pulsars. The peak of longitude at 0
or 360
implies that the pulsar population in the Galactic plane is concentrated in the region facing the Galactic Centre. A vast majority of these pulsars are observed in radio waveband and many have been detected in the optical, X-ray and high energy gamma-ray bands [
20].
On the basis of plausible physical processes involved in the emission of radiation, pulsars are classified in three types:
rotation powered pulsars,
accretion powered pulsars and
internal energy powered pulsars [
21]. In the rotation powered pulsars, emission of electromagnetic radiation is attributed to the loss of rotational kinetic energy of neutron star due to the magnetic braking. This group of pulsars are further classified as classical radio pulsars, millisecond radio pulsars, and mildly recycled pulsars depending upon the spin period (
P) and strength of neutron star surface magnetic field (
). The classical radio pulsars exhibit
1s and
G. Whereas millisecond radio pulsars and mildly recycled pulsars have
20s,
G and
20–100ms,
G respectively [
22]. In the accretion powered pulsars, material accretion on to the neutron star is responsible for emission of energetic radiation. These pulsars are generally observed as a binary system consisting of a neutron star as a compact object and a main sequence companion star as a donor star. Depending on the mass of companion star, the accretion powered pulsars are classified as low-mass X-ray binary and high-mass X-ray binary. In low-mass X-ray binary pulsars, the surface magnetic field of neutron star is weak (
G) due to extended phase of accretion [
23]. Thermonuclear X-ray burst, accretion-powered millisecond pulsations, kHz quasi-periodic oscillations, broad relativistic iron lines, and quiescent emission are the main physical processes taking place in low-mass X-ray binary systems. On the other hand, high-mass X-ray binaries harbour neutron stars with relatively strong surface magnetic field
. These objects are mostly observed in X-rays with hard spectrum [
24]. The X-ray emission from such pulsars has the signatures of Comptonisation due to interaction of accretion plasma with the magnetic field of neutron star and presence of Cyclotron broad absorption lines referred to as the Cyclotron Resonance Scattering Feature. Finally, the internal energy powered pulsars form a new class where exact physical mechanism for emission of radiation is not known. However, it is assumed that internal energy reservoir in the form of strong magnetic field decay or residual cooling powers the observed emission [
25]. Magnetars, soft gamma ray repeaters and anomalous X-ray pulsars are believed to be member of the group of internal energy powered pulsars [
26]. Another classification of known pulsars is also proposed on the basis of their measured spin period. According to this classification, pulsars are of two types:
millisecond pulsars (MSP) and
normal pulsars (NP). MSPs exhibit
20ms and
G, whereas NPs have
20ms and
G. It is believed that MSPs in binary systems are spun up by accreting matter and angular momentum from companion stars in a so called recycling process [
27].
The launch of large area telescope onboard the
Fermi satellite in 2008 has led to the discovery of a number of gamma-ray pulsars. However, only a few of them have been detected so far by the ground-based gamma-ray instruments. In this paper, we mainly focus on the gamma-ray pulsars and explore the potential of recently commissioned MACE (Major Atmospheric Cherenkov Experiment) telescope to detect the pulsed emission above 20GeV. The outline of paper is the following. In
Section 2, we discuss the empirical properties of pulsars. A brief review of gamma-ray pulsars is given in
Section 3. In
Section 4, we describe salient features of the MACE telescope. We discuss the results from the feasibilities studies for pulsar observations with MACE in
Section 5. Finally we summarize in
Section 6.
2. Pulsar Characteristics
A complete understanding of the underlying physical processes involved in the pulsed electromagnetic emission and geometry of the emission region is far from complete. Soon after the pulsar discovery, it was argued that, on the basis of observed features, the powerhouse of pulsar is a fast spinning/rotating neutron star with strong surface magnetic field [
11]. The periodic nature of pulsar emission observed on Earth, with incredible timing accuracy comparable to the atomic clock, is understood in the frame-work of a
lighthouse model as the magnetic axis of rotating neutron star sweeps across the line of sight of an observer. The emission mechanism is linked to the magnetic field affected co-rotating plasma in the pulsar magnetosphere with less understood structure and dynamics. It is still not clear whether the emission region is located out in the magnetosphere or close to the neutron star [
28,
29].
The spin-down law, given by Equation
1, is generally valid for any moment in the lifetime of the pulsars. Moreover, among the three parameters
I,
, and
, atleast two are expected to change for a real pulsar. Therefore,
suffers a slow decay due to the Ohmic dissipation of electric currents in the neutron star crust [
30]. The time evolution of obliquity angle
is given by
Measurements from radio observations suggest that
evolves on a timescale of 10
years [
31]. This can lead to a situation where
becomes zero and magnetic axis of an isolated pulsar eventually aligns with its rotation axis on a timescale of
. The moment of inertia
I varies slightly only during sudden small changes in the rotation velocity of pulsar, generally referred to as pulsar-glitches [
32]. This is attributed to the transfer of angular momentum between the crust and another component of the neutron star. The pulsar period
P also changes continuously for the rotation powererd pulsars due to the loss of kinetic energy. Therefore, most of the pulsars slow down over the time despite their
P being very stable and the slow-down rate is given by
. The parameters such as
P,
and higher order time-derivatives of
P are determined from the pulse time-of-arrival measurements over long intervals ranging from hours to decades [
33]. The precision in the measurement of the pulse time-of-arrival critically decides the amount of useful information that can be extracted for a pulsar observationally. The measured pulse time-of-arrival by a telescope on Earth corresponds to a non-inertial frame of reference measurement. Therefore, pulse time-of-arrivals are first translated to an inertial reference frame, normally the solar-system barycentre (an appropriate inertial reference frame), to remove the effects of Earth’s rotational and orbital motion. A barycentric pulse arrival time is estimated for individual time-of-arrival on the basis of a model for the intrinsic properties of the pulsar including its astrometric, rotational, propagation effects in interstellar medium and other parameters. The difference between estimated and observed pulase arrival times is referred to as the timing
residual and its study is very important for all pulsar timing measurements.
From the long-term pulsar timing measurements, the slow-down rate is derived as
where
is the pulse freqeuncy or spin-frequency of the pulsar. The intrinsic rotation of an isolated pulsar can be expressed by the Taylor-series expansion of
or
P about a reference time
as [
33]
or
where
T is pulsar proper time,
is the initial spin frequency at reference epoch
. The number of rotations completed by the neutron star from
is known as pulsar phase (
). The fractional part of
at a given epoch carries the information on orientation of pulsar at that time. Pulses with same fraction of
are measured when pulsar orientation is aligned with the observer. Distribution of
attributed to each event recorded by a detector/telescope is known as pulsar phase diagram or phasogram. This describes the pulsar rotation model in a reference frame comoving with the pulsar or pulsar-ephemeris which takes into account the changes in spin frequency with time due to the rotational energy loss. The pulsar spin down contributes to the time-derivatives of
or
. From Equation
4, pulse number
N can be written as [
34]
where
is the number of pulses at
. The pulsar proper time
T corresponds to its time at solar-system barycentre of infinite frequency up to a constant and is expresses as [
34]
where
is pulse arrival time at telescope,
is relative time correction, (
) is dispersion correction,
f is observation frequency,
is delay due to Earth-motion around Sun,
is the Shapiro delay due to space-time curvature, and
is the Einstein delay. For a pulsar in binary system, extra time delays caused by the orbital motion of neutron star and gravitational field of the companion star are also included in
T.
The distribution of
versus
P or
diagram is used to identify the underline physical process in the distinct group of pulsars [
35]. For rotation powered pulsars (i.e. NPs), values of
P (0.3-3 s) and
(
) are measured with high accuracy through the pulsar-timing measurements. For MSPs (mostly observed in binary systems),
P is mostly less than 20ms and
ranges from
to
. The smaller values of
for MSPs imply that their
P is more stable than that of NPs on a longer timescale. The knowledge of
P and
is also used to estimate the magnetic field at the pulsar surface as [
36]
assuming a magnetic-dipole field structure. However, the actual
could be much stronger due to presence of higher order magnetic poles at the neutron star surface. The spin-down history of an isolated pulsar can be captured by its
P and
. If the initial spin period
is much less than
P, (Equation
5), we have the
characteristic age of pulsar given by [
30]
This corresponds to the time since the pulse period was zero and is used to estimate the true pulsar age (
). The lines of constant
in the
diagram are known as death line. A simplified view of pulsars suggests that the evolution of radio pulsars proceeds along the line of constant magnetic field and descends towards the death line in the
diagram. After crossing the death line, a pulsar becomes radio quiet. The true age of a rotation powered pulsar is expressed as
where
is initial spin-down timescale and
n is the
braking-index which describes the secular spin down of pulsar. The braking-index is defined as [
37]
It is a characteristic tool to probe the magnetic field evolution.
3 and 5 correspond to the braking due to the magnetic-dipole and gravitational wave emission respectively. Braking by pulsar winds, which deform the magnetosphere, is characterized by the value of
n ranging between 1.0 to 3.0. It is also used to describe the spin-down of neutron star or pulsar using the torque equations as
where
K is a proportionality constant related to energy loss mechanism of pulsar [
38]. In case of a rotating magnetic dipole approximation, the spin-down luminosity (
) of pulsar is given by the rotation energy loss of the neutron star, i.e.
where
is the rotational kinetic energy of neutron star and
I is its moment of inertial [
4]. This expression can be further simplified as
and
where
R is the radius of neutron star. For magnetic dipole braking (
), Equation
14 simply reduces to
This suggests that the braking timescale, over which loss of rotational energy of neutron star will slow down the pulsar spin, is
years. Since
also depends on
, MSPs are more lumnious than NPs for a given surface magnetic field strength of the neutron star [
36].
Thermal emission from the surface of neutron star can be distinguished from the non-thermal emission arising from the pulsar magnetosphere in the X-ray band. Due to strong gravitational field of neutron star, general relativistic effects should be invoked in the interpretation of the observed X-ray luminosity of a pulsar [
39]. The apparent bolometric thermal luminosity measured by a distant observer is given by
where
is the thermal luminosity emitted from the surface of neutron star and
is the surface gravitational redshift. The apparent radii of neutron star is
and the atmosphere redshifted effective temperature is
For a canonical neutron star, 10km, , and , the general relativistic effects may be important because the distantly observed thermal luminosity will be larger than the intrinsic luminosity by a factor of 1.32 or 32% higher.
4. MACE Telescope
Major Atmospheric Cherenkov Experiment (MACE) is an extremely large state-of-the-art gamma-ray telescope located at an altitude of ∼4.3 km at Hanle, Ladakh in India. The telescope was commissioned in September 2021 after its successful first light from the standard candle Crab Nebula in April 2021 [
61]. Subsequently, various operating parameters of the telescope were optimized and tuned using a large number of engineering runs and repeated observations of the Crab Nebula. The telescope has been deployed for science data collection for more than one year. During this period, it has successfully detected high energy gamma-ray photons from a number candidate sources including the recent giant flaring episode of the radio galaxy NGC 1275 in December 2022 and January 2023 [
62,
63]. The MACE telescope is designed to cover the high energy gamma-ray band above 20 GeV with a point source sensitivity better than the existing stand-alone imaging atmospheric Cherenkov telescopes in the world [
64,
65]. At present, the telescope is capable of detecting the high energy gamma-ray photons from Crab Nebula with a sensitivity of 5
in a few minutes (internal communication) in the energy range above 50 GeV. A picture of the MACE telescope at Hanle is shown in
Figure 3. It is equipped with a big quasi-parabolic light reflector of 21 m diameter and 25 m focal length. This offers a total light collection area of ∼ 340
. Such a large area is achieved by using a tessellated structure comprising 356 mirror panels of size 0.986 m × 0.986 m each. Each panel consists of 4 square-shaped spherical metalic mirror facets of size 0.488 m × 0.488 m each [
66] with similar focal length. The reflecting surface of the mirror facets, coated with a protection layer of
of thickness ∼ 120 nm, has an average reflectance of ∼ 85%. An active mirror alignment control system employing linear actuators and diode-lasers fitted on each panel is used to obtain best possible parabolic approximation and optimum focusing in all orientations of the basket. An imaging camera at the focal plane has an optical field of view of
and a uniform pixel resolution of 0.125
. It consists of 1088 photo-multiplier tubes (PMTs) with high quantum efficiency (above 30%) and medium gain (∼10
). The MACE camera is custom-designed, event-driven system with state-of-the-art technologies for in-house signal processing electronics and high-speed data acquisition. The entire telescope structure is supported by an altitude-azimuth drive system. With telescope’s moving weight of 180 ton, the drive system enables pointing and tracking of a gamma-ray source with an accuracy of ∼ 1 arcmin in the presence of wind blowing at a speed of up to 45 km/h. A data archival storage system with a capacity of 80 TB is housed in the control room at the telescope site for safe and secure storage of the MACE data as a large data volume of ∼ 50 GB is generated during every hour of observation.
Combination of large light collector of the telescope and high altitude of the site offers a unique opportunity to achieve a low gamma-ray energy threshold close to ∼ 20 GeV of MACE. It helps in exploring the gamma-ray Universe beyond a redshift
[
67]. The low threshold energy of MACE makes it an ideal instrument for observation of gamma-ray pulsars detected by the
Fermi-LAT.
5. Pulsars with MACE
The gamma-ray spectra of most of the pulsars, observed by the Fermi-LAT, have cutoff energy in the range 1-10 GeV. Due to this, the gamma-ray flux at higher energy end of the spectrum decreases very fast. Therefore, ground-based telescopes with better sensitivity than Fermi-LAT in the energy range above 50 GeV play crucial role in determining the pulsar spectra in very high energy band. With this motivation, we have developed a formalism based on following arguments to select a sample of Fermi-LAT pulsars for their observation with the MACE telescope.
In general, pulsar light curves or phasograms exhibit two peak structure. The first peak position at phase
is attributed to the radio emission and is therefore referred to as main pulse. This is assigned zero phase and the position of second peak at phase
is calculated with respect to
. However, there exist pulsars like Geminga which have only one peak in their phasogram. The pulse amplitude corresponding to these peaks changes with the energy and
peak is observed to dominate with increasing energy. Therefore, observation of pulsars in very high energy range by the ground-based telescopes like MACE is governed by the detection of pulse at
at
confidence level in their phasogram over the huge isotropic cosmic ray background mainly due to protons. Thus, if a gamma-ray pulsar is residing within the nebula, both gamma-rays from nebula and cosmic ray protons will act as background for pulsar detection with ground-based telescopes. These background events lack in periodicity and hence no phase, whereas photons originating from the pulsars have well defined phase associated with their rotation. Thus, the background events are uniformly distributed in the light curve or phasogram of a pulsar and gamma-ray signal from a pulsar will pile up in a particular phase-bin. If
is the number of bins in the phasogram of a pulsar, then background will get reduced by a factor of
. For
30, the observation time required for detection of dominant gamma-ray peak (
) in the phasogram of a pulsar at 5
statistical significance level with the MACE telescope is estimated as
where
is the integral flux of dominating peak above 30 GeV in units of
. We apply this methodology to two well known gamma-ray pulsars Crab and Geminga for estimating the expected time for MACE observations as their gamma-ray spectra are available in the energy range covered by the MACE telescope.
5.1. Crab Pulsar
Crab pulsar (PSR J0534+2200) is the most studied pulsar in all energy bands from Radio to TeV and also the first pulsar detected by any ground-based gamma-ray telescope. It is normal pulsar with
33 ms and
. It is among the few pulsars that have been detected across the entire electromagnetic spectrum from radio up to gamma rays of beyond 1 TeV [
68]. Its emission profile has three components: two dominant pulses separated by 0.4 in observed from radio to TeV gamma-rays and third, bridge component. The main pulse corresponding to
at phase 0 dominates at radio frequencies. The second pulse corresponding to
at phase 0.4 (also referred to as interpulse) is weaker at radio frequnecies and becomes dominant at very high energy gamma-rays. The bridge component (defined as the pulse phase between the main pulse and the second pulse) is observed in gamma-rays up to 150 GeV [
68]. The stereoscopic MAGIC telescope has detected signal at 6.6
and 8.8
for
and
respectively in 152 hours of carb pulsar observations at zenith angles below 35
. The differential energy spectra of gamma-ray photons corresponding to
and
are described by a power law in the range 70 GeV to 1.5 TeV. The
spectrum is relatively hard and corresponding integral flux is two times that of
. For
, the corresponding detection time for MACE is estimated to be
61.38 hours using Equation
24.
5.2. Geminga
Geminga pulsar (PSR J0633+1746) is a radio-quiet gamma-ray pulsar with
237 ms and
. Located at a distance of 250 pc from Earth, it is one of the closest gamma-ray pulsar. The MAGIC telescope has detected pulsed gamm-ray signal from Geminga at 6.3
statistical significance level in 80 hours of observations [
69]. Only
peak is observed to be dominant in the phasogram in the energy range 15-75 GeV. The corresponding integral flux is estimated as
. For this flux level, the total observation time required for MACE observations in the energy range 30 GeV-5 TeV is calculated to be
144 hours.
5.3. Pulsar Catalog for MACE
As discussed in
Section 3.1, the current population of gamma-ray pulsars detected by the
Fermi-LAT is ∼ 300. The distribution of integral energy flux in the energy range 0.1–100 GeV for all the
Fermi-LAT pulsars is shown in
Figure 4. It is observed that majority of the pulsars have high energy flux
which is typically two order less than the corresponding crab pulsar flux. The
Fermi-LAT spectra of most of the pulsars are described by Equation
20. The cutoff energy can be approximately calculated as
The distribution of cutoff energy (
) for all the
Fermi-LAT pulsars is reported in
Figure 5. It is found that most of the pulsars have cut-off energy close to 0.8 GeV whereas only 33 pulsars (out of 292) have
2 GeV.
Applying the visibility criteria of the astrophysical sources from a given observatory, only 96 Fermi-LAT pulsars are found to be visible from the MACE site at hanle (32.8 N, 78.9 E). Next criteria is the identification of pulsars which have 2GeV. This constrain further reduces the number of pulsars from 96 to 8. At last, we check for the significance of pulsation in the energy range 10-30 GeV, overlapping with MACE, to be more than 2. Based on the above criterion, we are left with only 6 pulsar candidates (other than Crab and Geminga) from the 4FGL-DR3 catalog for MACE observations. A summary of these sources is given in the Table.