The power law relationship between the semi-major

axis and the product of the mass ratio and the

square of the rotation period ratio in an eccentric
system, its role and the physical mechanism of its
formation
Yongfeng Dai (  [email protected] )
Lishugou Physical Space Technology Company

Research Article

Keywords: Gravitation, solar-terrestrial relations, planets and satellites: formation, planet-star interactions,
stars: rotation

DOI: https://doi.org/10.21203/rs.3.rs-2404090/v1

License:   This work is licensed under a Creative Commons Attribution 4.0 International License.
Read Full License
MNRAS 000, 1–23 (20XX) Preprint 19 December 2022 Compiled using MNRAS LATEX style file v3.0

The power law relationship between the semi-major axis and the product
of the mass ratio and the square of the rotation period ratio in an eccentric
system, its role and the physical mechanism of its formation
Yongfeng Dai (戴永丰) 1★
1 Lishugou
Physical Space Technology Company
0168 Wangbaotai Village, Hongwei district, Liaoyang city, China

Accepted XXX. Received YYY; in original form ZZZ

ABSTRACT

This study introduces the concepts of mass ratio order of magnitude and synchronous orbit radius ratio, and investigates the
relationship between semi-major axis and synchronous orbit radius ratio in known systems with the same level of mass ratio
order of magnitude, the same level of spin-orbit angle, and eccentric/circular orbits, including binaries, star-planet systems and
planet-moon systems. Results show that in a system with an eccentric orbit (e>0.01) and a mass ratio order of magnitude greater
than −7.5217, the semi-major axis has a power-law distribution relationship with the product of the mass ratio and the square of
the rotation period ratio, and under conditions of spin-orbit alignment, the power-law exponent has a positive linear relationship
with the mass ratio order of magnitude and is greater than 0 and less than or equal to 1. Meanwhile, this study also concludes that
this power law relationship plays a decisive role in the formation of eccentric orbits. Based on Einstein’s equivalence principle
and Newton’s law of gravity, this study also proposes that a moving gravitational field may exert a drag force on a mass body
in the moving direction, and reveals the physical mechanism of the formation of this power law relationship. Based on such a
mechanism, this study proposes that for a spin-orbit misaligned system, its power-law exponent is close to the product of the
power-law exponent under conditions of spin-orbit alignment and the cosine of the spin-orbit angle.
Key words: Gravitation – solar-terrestrial relations – planets and satellites: formation – planet-star interactions – stars: rotation

1 INTRODUCTION of the rotation period ratio in an eccentric system lies in the intro-
duction of two system parameters: synchronous orbit radius ratio
The aim of this study is to determine whether there is a relationship
and mass ratio order of magnitude, and before investigating the re-
between the orbits of both components in an eccentric system and
lationship between synchronous orbit radius ratio and semi-major
their rotations. Some authors put forward the correlation between the
axis in known systems through regression analysis, these systems are
orbital period and spin period of pulsar in known high-mass X-ray
grouped according to their mass ratio orders of magnitude, spin-orbit
binaries (Corbet 1984; Li & van den Heuvel 1996), some authors
angles and eccentricities.
put forward the correlation between the eccentricity and spin period
For previous models that study the influences of rotation on
of pulsar in known double neutron star binaries (Dewi et al. 2005;
orbit, such as tidal torque (Witte & Savonije 2002), gravita-
Faulkner et al. 2005), and some authors also put forward the corre-
tional quadrupole moment (Campbell & Moffat 1983) and general-
lation between the orbital period and spin period of pulsar in known
relativistic frame-dragging (Ashby 2004), because they are only ap-
double neutron star binaries (Tauris et al. 2017). However, they only
plicable to short-period systems and their influences are weak, so
investigated the correlation between the orbit and rotation of one (i.e.
this power-law relationship could not be caused by these factors. In
pulsar) of the two components of known systems, but did not find
this work, based on Einstein’s equivalence principle and Newton’s
out the ways of investigating the relationship between the orbits of
law of gravity, we propose a physical mechanism of the formation of
both components and their rotations, so they did not give an exact
this power law relationship.
formula for such a relationship. In this work, we re-investigated such
The objects investigated in this study are binaries, star-planet sys-
correlations and found the investigation results support the conclu-
tems, star-asteroid systems and planet-moon systems; however, it is
sion that there is a power-law relationship between the semi-major
found that the situation of most exoplanets is different from that
axis and the product of the mass ratio and the square of the rotation
of planets in the solar system, for example, their orbital axis is not
period ratio in an eccentric system as proposed in this paper.
aligned with the rotation axis of their host stars (Triaud et al. 2010;
In this work, the discovery of a power-law relationship between
Winn et al. 2011; Albrecht et al. 2012), and their orbital period is less
the semi-major axis and the product of the mass ratio and the square
than the rotation period of their host stars (McQuillan et al. 2013;
Walkowicz & Basri 2013). Most authors believe that they formed
★ E-mail: [email protected] beyond the ice line at a few AUs and then migrated inward to their

© 20XX The Authors

2 Yongfeng Dai (戴永丰)
current orbits (Brewer et al. 2017; Cridland et al. 2019), where they simultaneously:
may have experienced a planet-planet scattering event (Dawson &
𝑚 2 𝑅22 (𝑣 1 sin 𝑖) 2
 3
Murray-Clay 2013; Bryan et al. 2016; Weiss et al. 2018). Therefore, 𝑟2
= . (2)
the planets we investigated do not include such exoplanets, even if 𝑟1 𝑚 1 𝑅12 (𝑣 2 sin 𝑖) 2
they have eccentric orbits. Surprisingly, however, we found that the
power-law relationship between the semi-major axis and the product
of the mass ratio and the square of the rotation period ratio in an 2.2 Division of eccentric systems and circular systems
eccentric system discovered through this work plays a decisive role
in the migration process of such planets. In this paper, we collectively refer to systems with e>0.01 as eccentric
The remaining sections of this work are organized as follows. In systems and systems with e<0.01 as circular systems, even if the
section 2, two system parameters are introduced, eccentric systems eccentricities of such circular systems are not close to 0 [which may
and circular systems are divided, and data of known systems is col- be related to the interaction between planets (or moons) because
lected and grouped. In section 3, a power law relationship between they exist in multi-planet systems (or multi-moon systems)]. Such
semi-major axis and synchronous orbit radius ratio in known eccen- a classification method is based on the fact that in this study, the
tric systems is discovered and its status in known circular systems is relationship between orbit and rotation exists in all those systems with
analyzed. In section 4, the decisive role of this power law relationship e>0.01 and mass ratio orders of magnitude greater than −7.5217, but
in the formation of eccentric orbits, the physical mechanism of the not in most systems with e<0.01.
formation of this power law relationship, and the relationship between
the exponent of this power law relationship and spin-orbit angle are
discussed. Finally, main conclusions of this work are summarized in 2.3 Grouping of known eccentric systems
section 5. We collected the data of eccentric systems with known semi-major
axes and synchronous orbit radius ratios, as shown in tables 1, 2 and
3, and grouped these systems according to the following principles:
(a) these systems are divided into five levels according to their mass
2 METHODS AND MATERIALS
ratio orders of magnitude, namely low (see the systems in group
2.1 Introduction of two system parameters (i), whose mass ratio orders of magnitude range from −10.9975
to −8.3006), low-moderate (see the systems in group (ii), whose
In this paper, the investigated systems refer to a system composed of
mass ratio orders of magnitude range from −6.7799 to −5.4969),
two components bound by gravity, such as binary systems, star-planet
moderate (see the systems in group (iii), whose mass ratio orders
systems, star-asteroid systems and planet-moon systems. Among the
of magnitude range from −4.3599 to −3.0202), moderate-high (see
two components, the one which has the larger mass is the primary,
the systems in group (vi), whose mass ratio orders of magnitude
and the other is the secondary; or the one which rotates faster is the
range from −1.9103 to −0.7984) and high (see the systems in groups
primary, and the other is the secondary, when the masses of the two
(iv), whose mass ratio orders of magnitude range from −0.1347 to
components are close to each other.
−0.0292, and the systems in groups (v), whose mass ratio orders of
In this paper, the mass ratio order of magnitude of a system refers
magnitude range from −0.4559 to −0.0167) levels; (b) the systems
to the logarithm of the ratio of mass of the secondary to that of the
with the same level of mass ratio order of magnitude, and the same
primary.
level of spin-orbit angle are included in the same group, even if
In this paper, the synchronous orbit radius ratio of a system refers
the masses of the systems in the same group may be of different
to the ratio of synchronous orbit radius of the secondary to that
orders of magnitude, for example, sun-giant planet systems and the
of the primary (the synchronous orbit radius of each component
Saturn-Titan system are included in group (iii); and (c) the binary
can be determined by such an equation: 𝑟 3 = 𝐺𝑚𝑃2 /4𝜋 2 , where,
stellar systems whose stellar radius and projected rotational velocity
𝑚 and 𝑃 represent the mass and rotation period of the component
data has been collected (see the systems in group (v)) will not be
respectively), as shown in the following equation:
included in the group which contains all the binary stellar systems
whose data on rotation periods has been collected (see the systems
2
𝑟 2 3 𝑚 2 𝑃2 in group (iv)) (because the rotation period of a star is obtained by
 
= , (1) analyzing asteroseismic data while the projected rotational velocity
𝑟1 𝑚 1 𝑃12
of a star is obtained by analyzing spectral data, and compared with
where 𝑟 1 and 𝑟 2 are the synchronous orbit radii of the primary and equation (1), the synchronous orbit radius ratio data determined by
the secondary, 𝑚 1 and 𝑚 2 are the masses of the primary and the equation (2) is also under the influence of the angle between the spin
secondary, and 𝑃1 and 𝑃2 are the rotation periods of the primary and axes of a system’s two stars and their stellar radii).
the secondary. All these eccentric systems have an eccentricity greater than 0.01,
When the rotation periods of the primary and secondary are un- and both of their components rotate super-synchronously. For group
known but their projected rotational velocities 𝑣 1 sin 𝑖 and 𝑣 2 sin 𝑖 (ii), the rotation of Iapetus in the Saturn-Iapetus system is judged
are known (because the two components of a binary stellar system to be slightly super-synchronous, just like Titan in the Saturn-Titan
can be traced back to the same portion of a molecular cloud, we can system; for group (iii), Titan in the Saturn-Titan system shows a
suggest that the the spin axis of the primary is aligned with that of the slightly super-synchronous rotation (Stiles et al. 2008), while the sun-
secondary) and their stellar radii 𝑅1 and 𝑅2 are given, then the syn- Uranus system, in which Uranus rotates sideways, is also classified
chronous orbit radius ratio of a binary stellar system can be obtained into this group; for group (vi), the moon in the Earth-moon system
by the following equation, which is obtained by substituting the ro- recedes from the Earth over time (Chapront et al. 2002), causing the
tation periods of the two components, expressed as 𝑃1 = 2𝜋𝑅1 /𝑣 1 moon’s orbital rate to become a bit slower than its rotation rate, so
and 𝑃2 = 2𝜋𝑅2 /𝑣 2 respectively, into equation (1) and multiplying the rotation of both components of the Earth-moon system, in which
the numerator and denominator of the resulting equation by sin2 𝑖 only the secondary rotates synchronously while the primary rotates

MNRAS 000, 1–23 (20XX)

A power law relationship in an eccentric system, its role and the physical mechanism of its formation 3
super-synchronously and spins down over time (Williams 2000), can derivative and spin period derivative data of known circular high-
be considered to be in a super-synchronous state. mass X-ray binary pulsars, as shown in table 10, and the orbital
Spin-orbit alignment is shown in these eccentric systems, except period derivative and spin period derivative data of pulsars in known
those in group (vi), in which spin-orbit misalignment is shown. As circular neutron star-white dwarf binaries and circular low-mass X-
we all know, these known systems in the solar system (exclusive of ray binaries, as shown in table 11.
the Earth-moon system) show a spin-orbit alignment; Ferdman et al.
(2013) found the spin-orbit angle of the double pulsar system PSR
J0737–3039 (see group (iv)) is less than 3.2 degrees; Albrecht et 3 RESULTS
al. (2011) found there is a spin-orbit alignment in the NY Cep (see
group (v)); Albrecht et al. (2013) proposed there is also a spin-orbit 3.1 Discovery of a power-law relationship
alignment in the EP Cru (see group (v)); because the two components By plotting the logarithms of the semi-major axes (the semi-major
of a binary stellar system can be traced back to the same portion of a axis is represented by 𝑎) of spin-orbit aligned systems with eccentric
molecular cloud, we can assume that there is spin-orbit alignment in orbits and a low-moderate-level mass ratio order of magnitude in
such binary stellar systems 61Cygni, 𝜉 Boo, 𝛼 Cen (see group (iv)), group (ii), spin-orbit aligned systems with eccentric orbits and a
𝛼 Sco, 𝛼 And, 𝛼 CrB, 𝛿 Ori and 𝛼 Vir (see group (v)). moderate-level mass ratio order of magnitude in group (iii), spin-
When collecting data from literature, we did not find relevant data orbit aligned systems with eccentric orbits and a high-level mass
on star-exoplanet systems and black hole binaries that meet the above ratio order of magnitude in group (iv) (as shown in table 1; see
requirements, nor select data on sun-asteroid systems with lower mass the text below for plotted data of the eccentric systems in group
ratio orders of magnitude. (i)), spin-orbit aligned systems with eccentric orbits and a high-level
In addition, from literature we also collected the spin period, orbital mass ratio order of magnitude in group (v) (as shown in table 2) and
period and eccentricity data of known eccentric high-mass X-ray bi- spin-orbit misaligned systems with eccentric orbits and a moderate-
nary pulsars (also including eccentric massive radio binary pulsars), high-level mass ratio order of magnitude in group (vi) (as shown
as shown in table 4, the orbital period derivative and spin period in table 3), against the logarithms of their synchronous orbit radius
derivative data of known eccentric high-mass X-ray binary pulsars ratios respectively, and carrying out linear regression respectively, as
(also including eccentric massive radio binary pulsars), as shown in shown in figures 1, we can obtain the following equations respectively
table 5, the spin period, orbital period and eccentricity data of the (given in the order of group (ii), group (iii), group (iv), group (v),
first-born pulsars in known eccentric double neutron star binaries, as group (vi)):
shown in table 6, and the orbital period derivative and spin period  
𝑟
derivative data of pulsars in known eccentric double neutron star log 𝑎 = −0.6570 log 2 + 9.3618, (3)
binaries, as shown in table 7. The eccentric millisecond pulsar PSR 𝑟1
J1903+0327 and its companion were once components of a hierar-
chical triple system (Freire et al. 2011), so data on this pulsar was
 
𝑟2
not selected, though we have known its relevant parameters. log 𝑎 = −1.4634 log + 8.8131, (4)
𝑟1

 
2.4 Grouping of known circular systems 𝑟2
log 𝑎 = −3.0073 log + 13.0390, (5)
𝑟1
We also collected the data of circular systems with known semi-major
axes and synchronous orbit radius ratios, which have an eccentricity  
less than 0.01 and show a spin-orbit alignment, as shown in tables 8 𝑟2
log 𝑎 = −1.4923 log + 10.3510, (6)
and 9. By referring to the above method about grouping known ec- 𝑟1
centric systems, we also grouped these known circular systems. For and
example, the systems in group (a) have a low-level mass ratio order  
𝑟2
of magnitude, the systems in group (b) have a low-moderate-level log 𝑎 = −0.6412 log + 8.7541. (7)
mass ratio order of magnitude, and the systems in group (c) have a 𝑟1
moderate-level mass ratio order of magnitude. Moreover, for the sys- These linear equations can be expressed in the form of a power-law
tems in these three groups, only the secondary rotates synchronously distribution:
while the primary rotates super-synchronously. In sun-Venus sys-   −𝐸
𝑟
tem, which has a low-moderate-level mass ratio order of magnitude, 𝑎=𝜑 2 , (8)
only the secondary rotates in a retrograde sense while the primary 𝑟1
rotates super-synchronously. In Pluto-Charon system, which has a where 𝐸 is a power-law exponent and 𝜑 is the coefficient.
high-level mass ratio order of magnitude, both components rotate From the above plot, we can conclude that there is a power-law
synchronously. In α Aur, which has a high-level mass ratio order of relationship between the synchronous orbit radius ratios and semi-
magnitude, only the secondary rotates synchronously while the pri- major axes of eccentric systems with the same level of mass ratio
mary rotates super-synchronously. In β Aur, which has a high-level order of magnitude (greater than or equal to the low-medium level)
mass ratio order of magnitude, the secondary has a rotation period and the same level of spin-orbit angle, and the power-law exponent
close to its orbital period. That is to say, only the secondary rotates (the negative of the slope of the above linear equation) has a depen-
synchronously while the primary rotates super-synchronously in this dence on the system’s mass ratio order of magnitude and spin-orbit
system. Torres, Claret and Young (2009) found there is a spin-orbit angle.
alignment in 𝛼 Aur. Because the two components of a binary stellar When the dependence of the power-law exponent on the mass
system can be traced back to the same portion of a molecular cloud, ratio order of magnitude of the above spin-orbit aligned systems is
we can assume that there is a spin-orbit alignment in 𝛽 Aur. investigated in the following, equation (6) will not be used, because
In addition, from literature we also collected the orbital period in the process of regression analysis to obtain this equation, the

MNRAS 000, 1–23 (20XX)

4 Yongfeng Dai (戴永丰)

13
11.5

11 12

10.5 11
log a

log a
10 10

9.5 9

9 8
-3.5 -2.5 -1.5 -0.5 0.5 -3 -2 -1 0
log(r2/r1) log(r2/r1)

(a) (b)

15 14
14
14
13 12
13
12
log a

12 10

log a
11
log a

10 11
8
9 10
8 9 6
0 0.5 1 1.5 -2.5 -1.5 -0.5 0.5 -6 -4 -2 0 2
log(r2/r1) log(r2/r1) log(r2/r1)

(c) (d) (e)

Figure 1. Linear regression between the logarithms of semi-major axes, log 𝑎, and the logarithms of synchronous orbit radius ratios, log(𝑟2 /𝑟1 ), for eccentric
systems in groups (ii), (iii) and (iv) in table 1, group (v) in table 2 and group (vi) in table 3. In figure 1(a), the solid square points from left to right are sun-Mars
system, sun-Earth system, sun-Mercury system and Saturn-Iapetus system respectively, which all have a low-moderate-level mass ratio order of magnitude and
present spin-orbit alignment. In figure 1(b), the solid square points from left to right are sun-Neptune system, sun-Uranus system, sun-Saturn system, sun-Jupiter
system and Saturn-Titan system respectively, which all have a moderate-level mass ratio order of magnitude and present spin-orbit alignment. In figure 1(c), the
solid square points from left to right are 61Cygni, 𝜉 Boo, 𝛼 Cen and PSR J0737–3039 respectively, which all have a high-level mass ratio order of magnitude and
present spin-orbit alignment. In figure 1(d), the solid square points from left to right are 𝛼 Sco, 𝛼 And, 𝛿 Ori, NY Cep, 𝛼 Vir, EP Cru and 𝛼 CrB respectively,
which all have a high-level mass ratio order of magnitude and present spin-orbit alignment. In figure 1(e), the solid square points from left to right are PSR
B1259–63/LS 2883, PSR J1740–3052/B-star, PSR J0045–7319/B-star and Earth-moon system respectively, which all have a moderate-high-level mass ratio
order of magnitude and present spin-orbit misalignment.

synchronous orbit radius ratio data used is obtained from equation magnitude as a continuous variable. Three levels of medians of mass
(2) using the projected rotational velocity, stellar radius and mass ratio orders of magnitude corresponding to equations (9), (10), and
data, and a lower goodness of fit value is obtained, while in the (5) are −6.11515 (determined by the data of the sun-Earth system and
process of regression analysis to obtain equations (3), (4) and (5), the the sun-Mercury system, as shown in table 1), −3.28205 (determined
synchronous orbit radius ratio data used is obtained from equation by the data of the sun-Jupiter system and the sun-Saturn system, as
(1) using the rotation period and mass data, and higher goodness of fit shown in table 1), and −0.08195 (determined by the data of PSR
values are obtained respectively; and equation (9) determined by the J0737–3039 and 𝜉 Boo, as shown in table 1) respectively from low
data of such sun-terrestrial planet systems as sun-Mercury system, to high.
sun-Earth system and sun-Mars system is used in place of equation The power-law exponents (i.e. the negatives of the slopes of equa-
(3) and equation (10) determined by the data of such sun-gas giant tions (9), (10) and (5)) are plotted against the medians of the mass
planet systems as sun-Jupiter system and sun-Saturn system is used ratio orders of magnitude at each level, as shown in figure 2. By
in place of equation (4) to remove the effects of other different kinds linear regression, a linear equation is obtained as below:
of systems (see below for an analysis of the effect of the quality of  
𝑚2
the original data used in the regression analysis in figure 1 on the 𝐸 = 0.4046 log + 3.0433. (11)
𝑚1
goodness of fit):
  By plotting the logarithms of the semi-major axes of spin-orbit
𝑟
log 𝑎 = −0.5513 log 2 + 9.6491, (9) aligned systems with eccentric orbits and a low-level mass ratio
𝑟1 order of magnitude in group (i) (as shown in table 1) against the log-
and arithms of their synchronous orbit radius ratios, as shown in figure 3,
  we can conclude that there is no power law relationship between the
𝑟2
log 𝑎 = −1.7214 log + 8.1064. (10) semi-major axis and synchronous orbit radius ratio in an eccentric
𝑟1
system with a low-level mass ratio order of magnitude. However, as
At the same time, we used the median (i.e. the average between the can be seen from figure 1, for an eccentric system whose semi-major
maximum and minimum values) of each level of mass ratio order of axis is in a power-law relationship with its synchronous orbit radius

MNRAS 000, 1–23 (20XX)

A power law relationship in an eccentric system, its role and the physical mechanism of its formation 5

4 worked out through equation (11) using the average value of their
mass ratio orders of magnitude. For such a system, therefore, its
3 power-law exponent is negatively related to its spin-orbit angle.
Among the original data of the 5 groups of eccentric systems used
in the linear regression in figure 1, the semi-major axis (or orbital
E

period) data, the mass data, which is used to calculate synchronous

1 orbit radius ratios, most of rotation period data (another parameter
used for calculating synchronous orbit radius ratios), and most of
0 spin-orbit angle data are of high quality. Most of these high-quality
-7 -6 -5 -4 -3 -2 -1 0
log(m2/m1)
data come from NASA’s Planetary Fact Sheets, on which no error
is given, so no error bars are provided in figures 1 and 2. In figure
Figure 2. Linear regression between the power-law exponents (i.e. the neg- 1(a), the 4 data points represent sun-Mars system, sun-Earth system,
atives of the slopes of equations (9), (10) and (5)), 𝐸, and the median at sun-Mercury system and Saturn-Iapetus system respectively, all of
each level of the mass ratio orders of magnitude, log(𝑚2 /𝑚1 ), of the systems which have eccentric orbits with a spin-orbit alignment and a low-
whose data is regressed to determine these equations. moderate-level mass ratio order of magnitude. In these systems, the
rotation period data of the sun, 1 gaseous planet, 3 solid planets and
1 solid moon is of high quality. And this plot shows a linear relation-
11.8
ship between the two parameters, with a coefficient of determination
11.6
11.4
of 0.996. In figure 1(b), the 5 data points represent sun-Neptune
11.2 system, sun-Uranus system, sun-Saturn system, sun-Jupiter system
11 and Saturn-Titan system respectively, all of which have eccentric
10.8 orbits with a spin-orbit alignment and a moderate-level mass ratio
log a

10.6
order of magnitude. In these systems, the rotation period data of
10.4
10.2
the sun, 3 gaseous planets and 1 solid moon is of high quality, the
10 gaseous planet Neptune, which is relatively farther away from us,
9.8 has a likely overestimated rotation period (gaseous planets and stars
-5 -4.5 -4 -3.5 -3 -2.5
present latitudinal differential rotation and radial differential rota-
log(r2/r1)
tion), and cause its data point to deviate from the regression line
(it may also have experienced small-degree planet-planet scattering
Figure 3. The logarithms of semi-major axes, log 𝑎, are plotted against the
logarithms of synchronous orbit radius ratios, log(𝑟2 /𝑟1 ), for eccentric sys- with Uranus, so it and Uranus are not the object of our investigation;
tems in group (i) in table 1. The solid square points from left to right are for more details, see the paragraph about planet-planet scattering
sun-Juno system, sun-Vesta system, sun-Pallas system, sun-Ceres system, analysis). And this plot shows a linear relationship between the two
Jupiter-Elara system and Jupiter-Himalia system respectively, which all have parameters, with a coefficient of determination of 0.9914. In figure
a low-level mass ratio order of magnitude and present spin-orbit alignment. 1(c), the 4 data points represent 61Cygni, 𝜉 Boo, 𝛼 Cen and PSR
J0737–3039 respectively, all of which have eccentric orbits with a
spin-orbit alignment and a high-level mass ratio order of magnitude.
ratio, its mass ratio order of magnitude is at low-moderate, moder- As binary stars, their stars’ rotation period data is of high quality,
ate, moderate-high or high-level. This indicates that for a system in and the rotation period data of the pulsars in PSR J0737–3039 is also
which there is such a power-law relationship, its mass ratio order of high quality. And this plot shows a linear relationship between
of magnitude is greater than a value between the low level (whose the two parameters, with a coefficient of determination of 0.9996. In
upper limit is −8.3006, as shown in table 1) and the low-moderate figure 1(d), the 7 data points represent 𝛼 Sco, 𝛼 And, 𝛿 Ori, NY Cep,
level (whose lower limit is −6.7799, as shown in table 1). On the 𝛼 Vir, EP Cru and 𝛼 CrB respectively, all of which have eccentric
other hand, when the power-law exponent (𝐸), an exponent of the orbits with a spin-orbit alignment (most of their spin-orbit angles are
power-law distribution, is greater than 0, the corresponding mass unknown) and a high-level mass ratio order of magnitude. Compared
ratio order of magnitude obtained by equation (11) is greater than to the previous group of binary stars, this group of binaries have un-
−7.5217. Therefore, for a system in which there is such a power- known rotation periods of the stars, but a range can be determined
law relationship, its mass ratio order of magnitude is higher than through stellar radii and projected rotation velocities. While there are
−7.5217. In addition, when the mass of the secondary of a spin-orbit still some data points scattered around the regression line, and this
aligned system is close to that of its primary, according to equation plot shows a linear relationship between the two parameters, with
(11), we can conclude that the power-law exponent (𝐸) of equation a coefficient of determination of 0.9592. In figure 1(e), the 4 data
(8) corresponding to this system should be equal to 3. The above points represent PSR B1259–63/LS 2883, PSR J1740–3052/B-star,
shows that the power-law exponent interval of equation (8) is greater PSR J0045–7319/B-star and the Earth-moon system respectively, all
than 0 and less than or equal to 3, and the corresponding mass ratio of which have eccentric orbits with a spin-orbit misalignment and
order of magnitude interval is greater than −7.5217 and less than or a moderate-high-level mass ratio order of magnitude. The rotation
equal to 0. period data of the Earth and the moon is of high quality, and the
Equation (7) is obtained by performing a regression analysis for the rotation period data of the pulsars of other systems is also of high
data of such spin-orbit misaligned systems as B1259–63/LS 2883, quality; however, the rotation periods of the B stars (also including
J1740–3052/B-star, J0045–7319/B-star and the Earth-moon system, LS 2883) are judged values given by the author in the literature, and
and the negative of its slope is 0.6412, which is the power-law expo- their spin-orbit angles are in a wide range (the power-law exponent
nent corresponding to these systems; if we assume there is spin-orbit is negatively related to the spin-orbit angle of the system, as can be
alignment in the above massive radio pulsar binaries, then the power- seen from the previous paragraph). While there are few data points
law exponent corresponding to these systems is 2.6241, which is scattered around the regression line, and this plot shows a linear

MNRAS 000, 1–23 (20XX)

6 Yongfeng Dai (戴永丰)
10000
relationship between the two parameters, with a coefficient of deter-
mination of 0.9823. In summary, in the linear regressions of the data
of the above 5 groups of eccentric systems, the reason for the slight 1000

decrease in the coefficient of determination of a few groups (see the

last two groups above) is lacking of high-quality rotation period data 100

Pb (d)
and spin-orbit angle data instead of any factors of the system itself.
Therefore, it is credible to conclude through the above linear regres- 10
sion that there is a power-law relationship between the semi-major
axis and the synchronous orbit radius ratio in an eccentric system 1
with a low-moderate, moderate, moderate-high, or high-level mass 0.01 0.1 1 10 100 1000 10000
P (s)
ratio order of magnitude.
Meanwhile, although there are only 3 data points in figure 2, they
Figure 4. Linear regression between orbital period, 𝑃𝑏 , and spin period,
represent 3 out of 4 levels of mass ratio orders of magnitude of all
𝑃, on a log-log scale for known eccentric high-mass X-ray binary pulsars
known eccentric systems that obey such a power-law relationship (also including eccentric massive radio binary pulsars) whose data has been
(only the data of those eccentric systems with a spin-orbit alignment collected into table 4.
and a moderate-high-level mass ratio order of magnitude is missed),
and these 3 data points all fall on a straight line (R2 = 1) and the data is
of high quality. In addition, the regression equation (11) is also proved the two parameters in equation (12) (in this equation, the semi-major
by the analysis of the mass ratio orders of magnitude interval and axis is equivalent to the orbital period, and the same is true for the
the power-law exponent interval of equation (8). Therefore, equation following) or (13). Although in these eccentric high-mass X-ray bi-
(11) is also believable. naries (also including eccentric massive radio pulsar binaries) with
By substituting equation (1) into equation (8), we can obtain: pulsars as secondaries, the masses of OB stars and pulsars are all
2 −𝐸/3
! within a narrow range, and in these eccentric double neutron star
𝑚 2 𝑃2 binaries with first born pulsars as primaries, the masses of secondary
𝑎=𝜑 × . (12)
𝑚 1 𝑃2 neutron stars and first born pulsars are all within a narrow range;
1
however, it can still be seen from equations (12) and (13) that the
Thus, we conclude that in a system with an eccentric orbit (e>0.01) differences of other parameters in different systems will affect the
and a low-moderate, moderate, moderate-high, or high-level mass relationship between the investigated parameters, and further lead
ratio order of magnitude (i.e. the mass ratio order of magnitude is to smaller coefficients of determination in regression analysis. Since
greater than −7.5217), there is a power-law distribution relationship only the correlation between two parameters will be investigated, no
between the semi-major axis and the product of the mass ratio and error bars were provided in the plots. Similarly, to support this con-
the square of the rotation period ratio, and the exponent (𝐸/3) of clusion, we will also investigate the correlation between the orbital
this power-law distribution has a positive linear relationship with the evolution and spin evolution of the pulsar in known eccentric high-
system’s mass ratio order of magnitude and is greater than 0 and mass X-ray binaries (also including known eccentric massive radio
less than or equal to 1 (under conditions of spin-orbit alignment) pulsar binaries) and known eccentric double neutron star binarie.
(see equation (11), where both slope and intercept are divided by For all known eccentric high-mass X-ray binary pulsars (also in-
3) and a negative correlation with the system’s spin-orbit angle. In cluding eccentric massive radio binary pulsars) whose data has been
addition, both components of these eccentric systems rotate super collected in table 4, their orbital periods are plotted on a log-log scale
synchronously (also including the case where its secondary rotates against their spin periods in figure 4, and the differences of 1 minus
synchronously while the primary rotates super-synchronously and their eccentricities are plotted on a log-log scale against their spin
spins down over time) periods in figure 5. For these pulsars, although different values of the
other parameters shown in equation (12) or (13) will greatly reduce
the coefficient of determination in regression analysis, it can still be
3.2 Support for such a power-law relationship seen that in figure 4, there is a negative correlation between the or-
Equation (12) can also be expressed as the relationship between the bital periods of these pulsars as the secondaries and their spin periods
eccentricity of an eccentric system and its rotation period ratio: (R2 = 0.3197), which is in line with the relationship between the two
parameters in equation (12), and in figure 5, there is a weak posi-
2 −𝐸/3
!
𝑎𝑝 𝑚 2 𝑃2 tive correlation between the differences of 1 minus eccentricities of
=𝜑 × . (13) these pulsars as the secondaries and their spin periods (R2 = 0.239),
1−𝑒 𝑚 1 𝑃2
1 which is in line with the relationship between the two parameters in
where 𝑎 𝑝 is the periastron separation. equation (13).
By investigating eccentric systems with known synchronous orbit Table 5 presents the collected data on known eccentric high-mass
radius ratios and semi-major axes we have concluded that in an eccen- X-ray binary pulsars (also including eccentric massive radio binary
tric system, there is a power-law relationship between the semi-major pulsars), mainly including orbital period derivative and spin period
axis and the product of the mass ratio and the square of the rotation derivative. Of them, PSR J1740–3052 and PSR J0045–7319 were
period ratio. However, the number of such known eccentric systems observed to undergo an orbital decay and spin-down, OAO 1657–
is not very large. To support this conclusion, we will expand our in- 415 was observed to undergo an orbital decay and a slight spin-up,
vestigation to the correlation between the orbital period/eccentricity EXO 2030+375 and Vela X–1 were observed to be unchanged in
and spin period of the first-born pulsar in known eccentric high-mass orbital period and a spin-up/spin-down close to zero (or a slight spin-
X-ray binaries (also including known eccentric massive radio pulsar up), and 2S 0114+650 and 4U 1538–52 were observed to have an
binaries) and known eccentric double neutron star binarie, and check increase in orbital period and a spin-up, over the whole investigation
whether this correlation is consistent with the relationship between period. The above phenomenon (exclusive of OAO 1657–415, whose

MNRAS 000, 1–23 (20XX)

A power law relationship in an eccentric system, its role and the physical mechanism of its formation 7
1 orbital periods of these first-born pulsars as the primaries and their
spin periods (R2 = 0.7318), which is in line with the relationship
between the two parameters in equation (12), and in figure 7, there is a
negative correlation between the differences of 1 minus eccentricities
0.1 of these first-born pulsars as the primaries and their spin periods (R2
1-e

= 0.3277), which is in line with the relationship between the two

parameters in equation (13).
For an eccentric system, by substituting Kepler’s third law 𝑎 3 =
0.01
𝐺 (𝑚 1 + 𝑚 2 ) 𝑃2𝑏 /4𝜋 2 into equation (12) and taking the time deriva-
0.01 0.1 1 10 100 1000 10000 tive of the resulting equation (assuming the components’ masses are
P (s)
constant over time), we can obtain the following equation:
 
Figure 5. Linear regression between the difference of 1 minus eccentricity, 𝐸/2 (𝐸−1) −𝐸 ¤ (−𝐸−1) ¤
2𝜋𝐸 𝜑3/2 𝑚 1 𝑃1 𝑃2 𝑃1 − 𝑃1𝐸 𝑃2 𝑃2
1 − 𝑒, and spin period, 𝑃, on a log-log scale for known eccentric high-mass 𝑃¤ 𝑏 = . (14)
𝐸/2
X-ray binary pulsars (also including eccentric massive radio binary pulsars) 𝐺 1/2 𝑚 2 (𝑚 1 + 𝑚 2 ) 1/2
whose data has been collected in table 4.
Table 7 presents the collected data on known eccentric double
neutron star binaries, mainly including the orbital period derivative
100 and spin period derivative of pulsar. It has been concluded in figure
1(c) that PSR J0737–3039 obeys equation (12), which is among
these eccentric double neutron star binaries, and its orbital period
10 derivative can be calculated by equation (14) using the masses, spin
periods and spin period derivatives of both components, as well as
Pb (d)

the power-law exponent and the coefficient (see equation (5)), i.e. the
1
orbital decay rate obtained is −5.29×10−12 , which is greater than the
observed value −1.25×10−12 as listed in table 7. This difference may
be associated with the change in mass of components and remains
0.1
   to be further studied. On the other hand, the pulsars in these systems
P (ms) are observed to have a high orbital decay rate and a significant spin-
down rate, and the spin down rate of the second-born pulsar as the
Figure 6. Linear regression between orbital period, 𝑃𝑏 , and spin period, 𝑃, secondary is several orders of magnitude greater than that of the first-
on a log-log scale for the first-born pulsars in known eccentric double neutron
born pulsar as the primary. The above phenomenon matches with the
star binaries whose data has been collected in table 6.
negative correlation between orbital period and spin period ratio in
equation (12).
1

3.3 Status of such a power law relationship in circular systems

By plotting figure 1(a) we have concluded that there is a power-law
relationship between semi-major axis and synchronous orbit radius
1-e

ratio in eccentric systems with a low-moderate-level mass ratio order

of magnitude and spin-orbit alignment, such as sun-Mercury system,
sun-Earth system, sun-Mars system and Saturn-Iapetus system. Since
0.1
the Sun-Venus system (see table 8), which has a circular orbit, also
   has a low-moderate-level mass ratio order of magnitude and spin-
P (ms) orbit alignment, we can use the logarithm of the semi-major axis and
the logarithm of the synchronous orbit radius ratio of the Sun-Venus
Figure 7. Linear regression between the difference of 1 minus eccentricity,
system as a data point (represented by a hollow square) in figure
1− 𝑒, and spin period, 𝑃, on a log-log scale for the first-born pulsars in known
1(a), and conclude that there is no power law relationship between
eccentric double neutron star binaries whose data has been collected in table
6. the semi-major axis and synchronous orbit radius ratio of the Sun-
Venus system, as shown in figure 8. On the other hand, the secondary
of the Sun-Venus system rotates in a retrograde sense.
abnormality may be related to the change in rotation period of its OB By plotting figure 1(c) we have concluded that there is a power-law
companion) matches with the negative correlation between the orbital relationship between semi-major axis and synchronous orbit radius
period and spin period of pulsar as secondary in equation (12). ratio in eccentric systems with a high-level mass ratio order of mag-
For the first-born pulsars in all known eccentric double neutron nitude and spin-orbit alignment, such as 61Cygni, 𝜉 Boo, 𝛼 Cen
star binaries whose data has been collected into table 6, their orbital and PSR J0737–3039. Since the Pluto-Charon system (see table 8),
periods are plotted on a log-log scale against their spin periods in which has a circular orbit, also has a high-level mass ratio order of
figure 6, and the differences of 1 minus their eccentricities are plotted magnitude and spin-orbit alignment, we can use the logarithm of the
on a log-log scale against their spin periods in figure 7. For the first- semi-major axis and the logarithm of the synchronous orbit radius
born pulsars in these systems, although different values of the other ratio of the Pluto-Charon system as a data point (represented by a
parameters shown in equation (12) or (13) will greatly reduce the hollow square) in figure 1(c), and conclude that there is no power
coefficient of determination in regression analysis, it can still be seen law relationship between the semi-major axis and synchronous orbit
that in figure 6, there is a strong positive correlation between the radius ratio of the Pluto-Charon system, as shown in figure 9. On

MNRAS 000, 1–23 (20XX)

8 Yongfeng Dai (戴永丰)
11.5 equation (12) is greater than their actual semi-major axis, and only
the secondary rotates synchronously while the primary rotates super-
11
synchronously.
10.5 By plotting figure 1(c) we have concluded that there is a power-law
relationship between semi-major axis and synchronous orbit radius
log a

10
ratio in eccentric systems with a high-level mass ratio order of mag-
9.5
nitude and spin-orbit alignment, such as 61Cygni, 𝜉 Boo, 𝛼 Cen and
PSR J0737–3039. Since 𝛼 Aur (see table 8), which has a circular
9 orbit, also has a high-level mass ratio order of magnitude and spin-
-3.5 -3 -2.5 -2 -1.5 -1 -0.5 0
log(r2/r1)
orbit alignment, we can use the logarithm of the semi-major axis and
the logarithm of the synchronous orbit radius ratio of 𝛼 Aur as a
Figure 8. The logarithms of semi-major axes, log 𝑎, are plotted against the data point (represented by a hollow square) in figure 1(c) (the syn-
logarithms of synchronous orbit radius ratios, log(𝑟2 /𝑟1 ), for eccentric sys- chronous orbit radius ratios of these 5 data points are all obtained by
tems in group (ii) in table 1 and the Sun-Venus system in table 8, which has a equation (1) using the rotation periods and masses of both compo-
circular orbit. The straight line is a linear regression of the data points (from nents), and conclude that there is a power law relationship between
left to right are sun-Mars system, sun-Earth system, sun-Mercury system and the semi-major axis and synchronous orbit radius ratio of 𝛼 Aur, as
Saturn-Iapetus system respectively) for these eccentric systems. The hollow shown in figure 11. On the other hand, the secondary of 𝛼 Aur ro-
square point is the Sun-Venus system. They all have a low-moderate-level tates synchronously while the primary rotates super-synchronously.
mass ratio order of magnitude and present spin-orbit alignment.
Therefore, based on equation (15), we can judge that its primary
rotates almost constantly over time.
14 By plotting figure 1(d) we have concluded that there is a power-law
13 relationship between semi-major axis and synchronous orbit radius
12 ratio in eccentric systems with a high-level mass ratio order of mag-
11 nitude and spin-orbit alignment, such as 𝛼 Sco, 𝛼 And, 𝛼 CrB, NY
10 Cep, EP Cru, 𝛿 Ori and 𝛼 Vir. Since 𝛽 Aur (see table 9), which has
log a

9 a circular orbit, also has a high-level mass ratio order of magnitude

8 and spin-orbit alignment, we can use the logarithm of the semi-major
7
axis and the logarithm of the synchronous orbit radius ratio of 𝛽 Aur
6
as a data point (represented by a hollow square) in figure 1(d) (the
-0.5 0 0.5 1 1.5 synchronous orbit radius ratios of these 8 data points are all obtained
log(r2/r1)
from equation (2) using the projected rotational velocities, stellar
radii and masses of both components), and conclude that there is a
Figure 9. The logarithms of semi-major axes, log 𝑎, are plotted against the power law relationship between the semi-major axis and synchronous
logarithms of synchronous orbit radius ratios, log(𝑟2 /𝑟1 ), for eccentric sys-
orbit radius ratio of 𝛽 Aur, as shown in figure 12. On the other hand,
tems in group (iv) in table 1 and the Pluto-Charon system in table 8, which has
a circular orbit. The straight line is a linear regression of the data points (from
the secondary of 𝛽 Aur rotates synchronously while the primary ro-
left to right are 61Cygni, 𝜉 Boo, 𝛼 Cen and PSR J0737–3039 respectively) tates super-synchronously. Therefore, based on equation (15), we can
for these eccentric systems. The hollow square point is the Pluto-Charon sys- judge that its primary rotates almost constantly over time.
tem. They all have a high-level mass ratio order of magnitude and present In order to determine the mass ratio order of magnitude interval
spin-orbit alignment. of a circular system which there is a power law relationship between
the semi-major axis and the synchronous orbit radius ratio and in
which only the secondary rotates synchronously while the primary
the other hand, both components of the Pluto-Charon system rotate
rotates super-synchronously, we also analyze the Earth-Moon system,
synchronously.
even if it has an eccentric orbit. For the Earth-Moon system, there
For a system whose semi-major axis is in a power law re-
is a power law relationship between the semi-major axis and the
lationship with its synchronous orbit radius ratio and only the
synchronous orbit radius ratio (see figure 1(e)), in which only the
secondary rotates synchronously while the primary rotates super-
secondary rotates synchronously while the primary rotates super-
synchronously, by replacing the orbital period in Kepler’s third law
synchronously and spins down over time (Williams 2000), so the
𝑎 3 = 𝐺 (𝑚 1 + 𝑚 2 ) 𝑃2𝑏 /4𝜋 2 with the rotation period of the secondary
semi-major axis determined by equation (15) increases over time,
and substituting this equation into equation (12) to eliminate the term
that is, the system should have an eccentric orbit. This is consistent
“the rotation period of the secondary”, we can obtain the following
with the fact the moon recedes from the Earth over time (Chapront
equation:
et al. 2002) and has an eccentric orbit (e>0.01). On the other hand,
𝜑3 𝐺 𝐸 (𝑚 1 + 𝑚 2 ) 𝐸 𝑚 1𝐸 𝑃12𝐸 the Earth-Moon system has a moderate-high-level mass ratio order
𝑎 3(𝐸+1) = . (15) of magnitude.
(2𝜋) 2𝐸 𝑚 2𝐸
Table 10 presents the collected data on known circular high-mass
By plotting the logarithms of the semi-major axes of spin-orbit X-ray binary pulsars, mainly including orbital period derivative and
aligned systems with a low, low-moderate or moderate-level mass spin period derivative. LMC X–4, Cen X–3 and SMC X–1 were
ratio order of magnitude and circular orbits in groups (a), (b) and observed to undergo an orbital decay and a slight spin-up throughout
(c) in table 8 against the logarithms of their synchronous orbit radius the whole investigation periods. The above phenomenon does not
ratios, as shown in figures 10, respectively, we can conclude that match with the negative correlation between the orbital period and
there is no power law relationship between semi-major axis and spin period of the pulsar as secondary in equation (12), so we can
synchronous orbit radius ratio in these circular systems. In these conclude that these circular high-mass X-ray binary pulsars do not
systems, on the other hand, their semi-major axis determined by obey equation (12). In addition, for these pulsars with short orbital

MNRAS 000, 1–23 (20XX)

A power law relationship in an eccentric system, its role and the physical mechanism of its formation 9

8.4 9.6
8.8

8.6 9.2
8
8.8

log a
log a

8.4

log a
7.6
8.2 8.4

7.2 8 8
-3.6 -3.1 -2.6 -2 -1.8 -1.6 -1.4 -1.2 -1 -1.4 -0.9 -0.4 0.1
log(r2/r1) log(r2/r1) log(r2/r1)

(a) (b) (c)

Figure 10. The logarithms of semi-major axes, log 𝑎, are plotted against the logarithms of synchronous orbit radius ratios, log(𝑟2 /𝑟1 ), for circular systems in
groups (a), (b) and (c) in table 8. In figure 10(a), the solid square points from left to right are Jupiter-Metis system, Saturn-Epimetheus system, Mars-Deimos
system, Jupiter-Amalthea system and Saturn-Janus system respectively, which all have a low-level mass ratio order of magnitude and present spin-orbit alignment.
In figure 10(b), the solid square points from left to right are Saturn-Enceladus system, Uranus-Miranda system, Saturn-Tethys system, Saturn-Dione system
and Saturn-Rhea system respectively, which all have a low-moderate-level mass ratio order of magnitude and present spin-orbit alignment. In figure 10(c),
the solid square points from left to right are Uranus-Ariel system, Uranus-Umbriel system, Jupiter-Io system, Jupiter-Europa system, Uranus-Titania system,
Uranus-Oberon system, Jupiter-Ganymede system and Jupiter-Callisto system respectively, which all have a moderate-level mass ratio order of magnitude and
present spin-orbit alignment.

14 periods, their primaries are likely to rotate synchronously or sub-

13 synchronously.
12
Table 11 presents the collected data on known circular neutron star-
11 white dwarf binaries and circular low-mass X-ray binaries, mainly
log a

including the orbital period derivative and spin period derivative of

10
the pulsar. In these circular binaries, the pulsars as primaries are
9 observed to spin up or down, whether they have an orbital decay
8
or an increase in orbital period, which reveals that in such systems,
0 0.5 1 1.5 the pulsar’s orbital evolution is not related to its spin evolution. On
log(r2/r1) the other hand, their semi-major axis determined by equation (12)
is shorter than their actual semi-major axis [because most of their
Figure 11. The logarithms of semi-major axes, log 𝑎, are plotted against primaries are millisecond pulsars, whose spin period is generally
the logarithms of synchronous orbit radius ratios, log(𝑟2 /𝑟1 ), for eccentric an order of magnitude lower than that of the pulsar as primary in
systems in group (iv) in table 1 and 𝛼 Aur in table 8, which has a circular
an eccentric double neutron star binary; as their secondaries the
orbit. The straight line is a linear regression of the data points (from left
to right are 61Cygni, 𝜉 Boo, 𝛼 Cen and PSR J0737–3039 respectively) for
white dwarfs or low-mass stars have a spin period from hours to
these eccentric systems. The hollow square point is 𝛼 Aur. They all have a days (or longer) (Kawaler 2004), which is much longer than the spin
high-level mass ratio order of magnitude and present spin-orbit alignment. period of the neutron star as secondary in an eccentric double neutron
star binary; and the mass of such a white dwarf or star is slightly
smaller than that of a neutron star]. Therefore, we can conclude that
15
these circular systems do not obey equation (12). In these systems,
14 on the other hand, only the secondary rotates synchronously while
the primary rotates super-synchronously (or both components rotate
13
super-synchronously), and these sytemns have a moderate-high, or
log a

12 high-level mass ratio order of magnitude.

11
Based on the above analysis, we conclude that there is no power
10 law relationship between semi-major axis and the product of mass
9 ratio and the square of rotation period ratio in most systems with
-2.5 -2 -1.5 -1 -0.5 0 0.5 circular orbits (e<0.01); a system may have circular orbits in one
log(r2/r1)
of the following circumstances: the semi-major axis of a system
Figure 12. The logarithms of semi-major axes, log 𝑎, are plotted against
in which both components rotate super-synchronously (or only the
the logarithms of synchronous orbit radius ratios, log(𝑟2 /𝑟1 ), for eccentric secondary rotates synchronously while the primary rotates super-
systems in group (v) in table 2 and 𝛽 Aur in table 9, which has a circular orbit. synchronously) determined by equation (12) is smaller than its actual
The straight line is a linear regression of the data points (from left to right semi-major axis, a system with a mass ratio order of magnitude is
are 𝛼 Sco, 𝛼 And, 𝛿 Ori, NY Cep, 𝛼 Vir, EP Cru and 𝛼 CrB respectively) smaller than that of the Earth-moon system and in which only the
for these eccentric systems. The hollow square point is 𝛽 Aur. They all have secondary rotates synchronously while the primary rotates super-
a high-level mass ratio order of magnitude and present spin-orbit alignment. synchronously, both components rotate synchronously, the secondary
rotates in a retrograde sense, and primary rotates synchronously or
sub-synchronously; for a circular system, only when its secondary
rotates synchronously while the primary rotates super-synchronously
and almost constantly over time, and its mass ratio order of magnitude

MNRAS 000, 1–23 (20XX)

10 Yongfeng Dai (戴永丰)
is equal to that of the Earth-moon system or higher, there may be
such a power law relationship in this system. e.g. 𝛼 Aur and 𝛽 Aur. The outer planet

The inner planet

4 DISCUSSION
The star
4.1 The decisive role of such a power law relationship in the
formation of eccentric orbits
Figure 13. Schematic diagram of a close encounter between inner and outer
It is generally believed that a massive radio pulsar binary (or a high- planets during migration. On the inner planet, the short arrow is the direction
mass X-ray binary) is generated by a massive close binary through along which this planet is accelerated and orbits, with the star as a reference,
supernova explosion, which results in the birth of a neutron star in and the long arrow is the direction along which this planet is accelerated, with
the system. The spin rate of the collapsed core increases rapidly the two planets’ common center of mass as a reference. On the outer planet,
during the collapse of the stellar core, so based on equation (12), the short arrow is the direction along which this planet orbits, with the star
we can propose that part of the collapsed core’s spin kinetic energy as a reference, and the long arrow is the direction along which this planet is
is exported and transformed into the orbital energy of both itself accelerated, with the two planets’ common center of mass as a reference. The
black dot represents the two planets’ common center of mass.
and its companion, thus new orbits are formed. On the other hand,
the spin axes of the two components of a close binary are generally
aligned with each other (Howe & Clarke 2009). Therefore, we can
propose that the spin-orbit misalignment in massive radio pulsar migrating planets. When a very close encounter occurs between two
binaries (also including high-mass X-ray binaries) is caused by a planets, if the semi-major axis of the system composed of these two
random tilt of the spin axis of the collapsed core against that of its planets determined by equation (12) is much larger than the distance
companion during the collapse of the stellar core and turns into a between them, then these two planets will scatter, that is, they would
neutron star in their predecessor. If the spin axis does not deviate in be accelerated in the direction of their spin around their common
this process, then the semi-major axis of such a system can reach up to center of mass, causing the outer planet to be ejected and become
1.3×1021 AUs when calculated by equation (12) using the parameters a distant planet or another planet, and the inner planet slows down
of PSR B1259–63/LS 2883 (see Table 3) (the power-law exponent rapidly, or even retrograde because the inner planet is accelerated
is determined by equation (11), the coefficient being the intercept of around the common center of mass of the two planets in the direction
equation (7)). This demonstrates that in this process, the collapsed opposite to that of the inner planet around the host star, as shown in
core and its companion may break away from the gravitational bond figure 13, and quickly falls down to its host star, which may cause
between them and form a single neutron star and a runaway OB its spin axis and orbital axis to deflect (such an axial deflection may
star after being accelerated in the direction of the collapsed core’s occur as well when there is another planet existing on the inner side
spin around their common center of mass. For example, it has been of this inner planet and the semi major axis of the system composed
found that single pulsars (i.e. visible single neutron stars) have an of these two inner planets, determined by equation (12), is greater
initial velocity of 320–420 km s−1 (Faucher-Giguère & Kaspi 2006) than the distance between them), and be captured by its host star
and runaway OB stars have an initial velocity higher than 25 km in a variety of orbits and becomes a hot Jupiter or another planet.
s−1 (Portegies Zwart 2000). These two velocities are more like the Therefore, planets that have experienced planet-planet scattering are
orbital velocities of two components in a massive radio pulsar binary. likely not to obey equation (12), even if they have eccentric orbits.
Therefore, we can propose it is the rapid increase in spin rate of the Planet-planet scattering may also occur to a lesser degree, such as
collapsed core that causes itself and its companion to be accelerated between Earth and Venus, and between Neptune and Uranus, where
by a pair of forces acting on them in the spin direction of the collapsed a small degree of planet-planet scattering event may have occurred,
core, which are equal in magnitude and opposite in direction, because resulting in a deflected axis of spin of the inner planet.
in a system, the product of the orbital velocity of one component and During the above-mentioned migration of multiple planets, when
its mass is equal to the product of the orbital velocity of the other a close encounter occurs between two planets, if the masses of these
and its mass. two planets are close to each other or one of the planets is slightly
It is generally believed that in a protoplanetary disk, a planetary smaller, then these two planets cannot form a stable system even if
embryo orbits at a velocity lower than the Kepler orbital velocity such a planet-planet system’s semi-major axis determined by equa-
due to the presence of significant drags from materials, causing it to tion (12) is less than, equal to, or slightly larger than the distance
migrate inward and thus gain a growth of mass, when it grows up to between them, because the semi-major axis of the system composed
the same mass as a planet, it will migrate outwards spontaneously of such a planet-planet system and the host star, determined by equa-
(Cossou et al. 2014; Muley et al. 2019). The reason for this sponta- tion (12), is smaller than that of the system composed of each planet
neous outward migration should be when the logarithm of the ratio and its host star determined by equation (12) because the orbital
of the mass of a newborn planet to that of its host star is greater than period of such a planet-planet system, that is, the system’s rotation
−7.5217, if its semi-major axis determined by equation (12) is greater period, is usually greater than the rotation period of each planet.
than the distance between it and its host star, the orbital separation Instead, these two planets are accelerated in the direction of their
between this planet and its host star will obey equation (12), that is, spin around their common center of mass, causing the outer planet
as the planet’s rotation rate increases due to its mass accretion, this to migrate outwards, and the inner planet migrates inwards because
planet will be accelerated in its rotating direction around its host star the inner planet is accelerated around the common center of mass of
(which is approximately the system’s center of mass) and migrate the two planets in the direction opposite to that of the inner planet
outwards into a spiral orbit and continue moving until it moves in an around the host star, as shown in figure 13, until each planet forms a
eccentric orbit when no such mass accretion rates appear. stable system with its host star.
In the above-mentioned protoplanetary disk, there may be multiple During the above-mentioned migration of multiple planets, when

MNRAS 000, 1–23 (20XX)

A power law relationship in an eccentric system, its role and the physical mechanism of its formation 11
a close (or a very close) encounter occurs between a large planet and that 𝑚 2 in this formula is the total number of physical space beams
a small planet (or several small planets), if the logarithm of the ratio representing the secondary’s gravitational field (in order to better
of the mass of the small one to that of its host star is less, equal to, or describe the distribution of gravitational field, the concept of phys-
slightly greater than −7.5217 and the two planets’ common center of ical space beams is introduced here), and such beams are radially
mass is approximately inside the large one, and the semi-major axis distributed from the secondary to infinite, and any segment of such a
of the system composed of the two planets determined by equation physical space beam is a local Minkowski space that moves together
(12) is slightly longer, equal to, or shorter than the distance between with a free fall at an acceleration of 4𝜋𝐺 towards the secondary, then
them (or this system does not obey equation (12)), then these two 𝑚 2 /4𝜋𝑎 2 is the number of such local Minkowski spaces per unit area
planets would be accelerated in the direction of their spin around their of the surface of this sphere, and (4𝜋𝐺)(𝑚 2 /4𝜋𝑎 2 ), the the gravi-
common center of mass to form a planet-moon (or moons) system. tational acceleration towards the secondary, is equal to the sum of
In such a star-planet-moon system, the orbital separation between the the accelerations of the free falls that move together with these local
planet and the host star obeys equation (12), and neither the orbital Minkowski spaces per unit area of the surface of this sphere (this
separation between the moon and the host star nor that between the means that the relationship between two adjacent local Minkowski
planet-moon system and the host star obey equation (12) because the spaces in these local Minkowski spaces per unit area is one local
orbital separation between the planet-moon system and the host star Minkowski space being located in another local Minkowski space).
determined by equation (12) is shorter than that between the planet By reference to this analysis result, we can propose that in the local
and the host star determined by equation (12) and the logarithm of physical space where the primary is located, the acceleration (which
the ratio of the mass of the moon to that of the host star is less, equal is derived from Einstein’s equivalence principle) generated by the
to, or slightly greater than −7.5217. rotation of the secondary’s gravitational field is equal to the product
If the mass of an outwardly migrating planet is large enough, such of the free-fall acceleration in the rotational direction from an initial
as a brown dwarf, then the system composed of this planet and its velocity of zero to the current velocity of the gravitational field that
host star acts as the primary and forms a hierarchical triple system moves together with a local Minkowski space (such a free fall accel-
together with the outer planet. Obviously, the orbital radius of the eration is equivalent to the current velocity of the gravitational field,
outer planet determined by equation (12) is very large, which means i.e. 2𝜋𝑎/𝑃2 ) and the number of such local Minkowski spaces per unit
that such a massive planet is impossible to have a close encounter area (perpendicular to the above sphere surface). On the other hand,
with the outer planet and that the probability of survival for such an as shown in figure 14, in this local physical space, the number of lo-
outer planet is very small. cal Minkowski spaces per unit area while moving together with free
falls in the direction of the gravitational field’s rotation is equal to
the square root of the number of local Minkowski spaces on another
4.2 The physical mechanism of the formation of such a power unit area (perpendicular to the previous unit area) while moving to-
law relationship gether with free falls towards the secondary, that is, (𝑚 2 /4𝜋𝑎 2 ) 1/2 .
Einstein’s equivalence principle states that in a local physical space Therefore, in the local physical space where the primary is located,
of a gravitational field, the equivalence between acceleration and the magnitude of the acceleration (which is derived from Einstein’s
gravity requires that this local physical space be regarded as a local equivalence principle) generated by the rotation of the secondary’s
Minkowski space (a space-time with three space dimensions and one gravitational field is equivalent to the product of the current velocity
time dimension) that moves together with a uniformly accelerated of the gravitational field and the square root of the number of local
free-fall towards the host. Thinking about this regard reversely, if Minkowski spaces per unit area while moving together with free falls
there is a local Minkowski space that moves together with a uniformly towards the secondary, that is, (2𝜋𝑎/𝑃2 )(𝑚 2 /4𝜋𝑎 2 ) 1/2 . Therefore,
accelerated free-fall in a local physical space, then there should also the force acting on the primary in the rotating direction caused by
be a force in the direction of the free-fall in this local physical space the rotation of the secondary’s gravitational field is
(if there is a mass body in this local physical space). If we assume 
2𝜋𝑎

𝑚 2 1/2

a rotating body’s gravitational field rotates together with it (see the 𝑓2 = 𝑚1 . (16)
𝑃2 4𝜋𝑎 2
second paragraph below for the discussion on this assumption), then
we can suggest that a local physical space of this rotating body’s Similarly, by analyzing 𝐹1 = (4𝜋𝐺)(𝑚 1 /4𝜋𝑎 2 )𝑚 2 , we have con-
gravitational field can be regarded as a local Minkowski space that cluded that the force acting on the secondary in the rotating direction
moves together with a uniformly accelerated free-fall in the direction caused by the rotation of the primary’s gravitational field is
of the gravitational field’s rotation from an initial velocity of zero
𝑚 1 1/2
  
2𝜋𝑎
to the current velocity of the gravitational field. Therefore, in this 𝑓1 = 𝑚2 . (17)
local physical space, there is a gravitational force pointing to the 𝑃1 4𝜋𝑎 2
rotating body, at the same time, there also exists a force in the rotating In addition, based on the fact that gravitational field is a physical
direction of the gravitational field caused by the gravitational field’s space and can be understood as a form of beams radially distributed
rotation (if there is a mass body in this local physical space). from their host to infinite, and combined with the fact that the pre-
For a system in which both components rotate super-synchronously dictions of general relativity for the effects of gravitational fields
and two spin axes are aligned with each other, we can express New- have been confirmed, we can suggest that for a rotating body, its
ton’s law of gravity as 𝐹2 = (4𝜋𝐺)(𝑚 2 /4𝜋𝑎 2 )𝑚 1 , which is a gravi- gravitational field should also rotate together with it.
tational force acting on the primary with a mass of 𝑚 1 and a rotation Through grouping known systems, we have known that a system
period of 𝑃1 , and 𝐹1 = (4𝜋𝐺)(𝑚 1 /4𝜋𝑎 2 )𝑚 2 , which is a gravitational in which only the primary rotates super-synchronously usually has
force acting on the secondary with a mass of 𝑚 2 and a rotation pe- circular orbits (here we take the center of mass of the system as ref-
riod of 𝑃2 (𝑎 is the distance between the primary and the secondary). erence), while both components of an eccentric system rotate super-
First, let’s analyze 𝐹2 = (4𝜋𝐺)(𝑚 2 /4𝜋𝑎 2 )𝑚 1 , where 4𝜋𝑎 2 is the synchronously and usually has an orbital precession. Therefore, such
surface area of a sphere with a radius of 𝑎 centered on the secondary circular orbits can be regarded as the result of the force acting on the
and this surface passes through the primary. If we can understand secondary caused by the rotation of the primary’s gravitational field,

MNRAS 000, 1–23 (20XX)

12 Yongfeng Dai (戴永丰)
1 unit
perpendicular to the drag force (and the reaction force against the
other drag force) acting on this component, so by referring to the

nit
Free falls in the direction momentum theorem we can establish the relationship between such

1u
of the gravitational
field’s rotation a drag force and the circular orbit velocity caused by it, and that
between such a reaction force and the circular orbit velocity caused

1 unit
by it, so as to determine the circular orbit velocity that is converted
into the semi major axis. At the same time, we select the value of
the time needed to form each circular orbit velocity as constant 𝑡,
Free falls towards because the magnitude of each circular orbit velocity has nothing to
the rotating body
do with that of such a force.
Taking the secondary as a reference, if the drag force acting on
the secondary caused by the rotation of the primary’s gravitational
field is represented by 𝑓1 (which also represents the reaction force
Figure 14. A schematic diagram of the relationship between the number of acting on the primary against this drag force) and the circular orbit
local Minkowski spaces (e.g. 3) per unit area of a local physical space on the velocity of the primary caused by this reaction force is represented by
equatorial plane of a rotating body while moving together with free falls in the 𝑣 𝑏1 , and the drag force acting on the primary caused by the rotation
gravitational field’s rotational direction and the number of local Minkowski of the secondary’s gravitational field is represented by 𝑓2 and the
spaces (e.g. 9) on another unit area (perpendicular to the previous unit area) circular orbit velocity of the primary caused by it is represented
of this local physical space while moving together with free falls towards the by 𝑣 𝑏2 , when an eccentric orbit is formed in a system due to the
rotating body. Each square prism represents a local Minkowski space. simultaneous existence of such two drag forces, then 𝑓1 𝑡 = 𝑚 1 𝑣 𝑏1
and 𝑓2 𝑡 = 𝑚 1 𝑣 𝑏2 , and 𝑣 𝑏2 = ( 𝑓2 / 𝑓1 )𝑣 𝑏1 is obtained by combining
these two formulas. No matter how long the distance between the two
which also means no matter how long the distance between the two components is, 𝑣 𝑏1 is always the circular orbit velocity, which will
components is and how big this force is, such a force will result in not lead to an increase in the distance between the two components,
the same result, that is, circular orbits, and such eccentric orbits and so it can be regarded as a constant 𝑐, so
their precession can be regarded as the result of joint action of the  
𝑓
force acting on the secondary caused by the rotation of the primary’s 𝑣 𝑏2 = 𝑐 2 . (18)
𝑓1
gravitational field and the force acting on the primary caused by the
rotation of the secondary’s gravitational field, which also means that On the other hand, however, it has been found that low mass ra-
in the presence of the circular orbit velocity (here we take one of tio systems preferentially have low eccentricities (Kane et al. 2012;
the two components as reference) caused by the force acting on the Bowler et al 2020). 𝑣 𝑏2 leads to the formation of the system’s semi-
secondary due to the rotation of the primary’s gravitational field, the major axis as well as an increase in eccentricity, so 𝑣 𝑏2 is constrained
orbital velocity caused by the force acting on the primary due to the by the ratio of the orbital radius of the primary to that of the sec-
rotation of the secondary’s gravitational field will be converted into ondary (i.e. mass ratio), that is, the smaller the ratio, the stronger
the distance between the two components (semi major axis). the constraint is, and the smaller 𝑣 𝑏2 . On the other hand, when there
During the formation of circular orbits described as above, because is 𝑣 𝑏2 only, such a constraint enables 𝑣 𝑏2 not to reach the circular
the rotation velocity of the primary’s gravitational field at the position orbit velocity unless the two orbital radii are close to each other.
of the secondary is greater than the orbital velocity of the secondary, Therefore, equation (18) is only applicable to those binaries whose
and the force acting on the secondary caused by the rotation of the two orbital radii are close to each other, which is also a power law
primary’s gravitational field is in direct proportion to the rotation relationship with an exponent of 1. For systems in which the primary
velocity of the gravitational field (as shown in equation (17)), and has an orbital radius of almost zero, for example, in a sun-asteroid
the primary also gains an orbital velocity due to this force, we can system, under the constraint of such a ratio, 𝑣 𝑏2 is nearly zero. The
conclude that the force exerted on the secondary by the rotation of exponent of the power law relationship applicable to such a system
the primary’s gravitational field is of a drag force, which leads to shall be 0. To sum up, the relationship between 𝑣 𝑏2 and ( 𝑓2 / 𝑓1 ),
a reaction force acting on the primary, and the two forces, which which is constrained by the ratio of the orbital radius of the primary
are equal in magnitude and opposite in direction, takes the system’s to that of the secondary, is a power law relationship with an exponent
mass center as axis (this also means that gravitational field is rigid). ranging from more than 0 to less than or equal to 1 (the exponent is
Similarly, a drag force is also exerted on the primary by the rotation of represented by 𝑥), that is,
the secondary’s gravitational field, leading to a reaction force acting  𝑥
𝑓
on the secondary, which takes the system’s mass center as axis, and 𝑣 𝑏2 = 𝑐 2 . (19)
𝑓1
the result is also that they move in circular orbits, if there is only this
pair of forces in the system. Therefore, based on equation (19), we can transform equation (12)
Due to the existence of a gravitational force perpendicular to the into equation (20) as below, which expresses the power law relation-
drag force caused by the rotation of the gravitational field, a mass ship between the orbital velocity caused by the drag force acting on
body may have a circular free fall trajectory in the rotating direction the primary due to the rotation of the secondary’s gravitational field
of the gravitational field under the action of such a drag force, if we and the ratio of the two drag forces:
take the host of the gravitational field as the reference. That means " # 𝐸/3
2 1/2
that the magnitude of this circular orbit velocity has nothing to do 1/2 1/2 𝑚 1 (2𝜋𝑎/𝑃2 ) (𝑚 2 /4𝜋𝑎 )
𝑎 =𝜑 , (20)
with that of this drag force. However, when there are two such drag 𝑚 2 (2𝜋𝑎/𝑃1 ) (𝑚 1 /4𝜋𝑎 2 ) 1/2
forces existing in the same system, each component suffers from
a drag force and the reaction force of the other drag force at the where 𝑚 1 (2𝜋𝑎/𝑃2 )(𝑚 2 /4𝜋𝑎 2 ) 1/2 is the drag force acting on the
same time, and the gravitational force acting on this component is primary caused by the rotation of the secondary’s gravitational field

MNRAS 000, 1–23 (20XX)

A power law relationship in an eccentric system, its role and the physical mechanism of its formation 13
(see equation (16)), 𝑚 2 (2𝜋𝑎/𝑃1 )(𝑚 1 /4𝜋𝑎 2 ) 1/2 is the drag force it is represented by 𝑣 𝑏𝑐1 , and the time needed to form this circular or-
acting on the secondary caused by the rotation of the primary’s bit velocity is represented by 𝑡; when 𝜃 is greater than 0 but less than
gravitational field (see equation (17)), 𝐸/3 is determined by the ratio 90 degrees, the drag force acting on the secondary caused by the rota-
of the orbital radius of the primary to that of the secondary (i.e. mass tion of the primary’s gravitational field in the direction of the reaction
ratio) (it has a minimum value 0 for systems in which the primary has force acting on the secondary is represented by 𝑓1 , the orbital velocity
an orbital radius of almost zero and a maximum value 1 for binaries caused by it is represented by 𝑣 𝑏1 , and the time needed to form this
whose two orbital radii are close to each other), and 𝑎 1/2 is equivalent orbital velocity is also represented by 𝑡 because 𝑓1 is a component
to 𝑣 𝑏2 , that is, the orbital velocity caused by the drag force acting on force of 𝑓𝑐1 . Since these two forces both act on one component, and
the primary due to the rotation of the secondary’s gravitational field. the gravitational force acting on this component is perpendicular to
the drag force acting on this component, we can establish the rela-
tionship between each of such forces and the orbital velocity caused
4.3 The relationship between the power law exponent and the by it by referring to the momentum theorem, so as to determine the
spin-orbit angle relationship between these two orbital velocities, expressed respec-
During the formation of a system’s eccentricity orbits, if the sec- tively as 𝑓𝑐1 𝑡 = 𝑚 2 𝑣 𝑏𝑐1 , where 𝑓𝑐1 = (2𝜋𝑎/𝑃1 )(𝑚 1 /4𝜋𝑎 2 ) 1/2 𝑚 2 ,
ondary’s spin axis is tilted at an angle (represented by 𝜃) of less than and 𝑓1 𝑡 = 𝑚 2 𝑣 𝑏1 , where 𝑓1 ≈ (2𝜋𝑎/𝑃1 ) cos 𝜃 (𝑚 1 /4𝜋𝑎 2 ) 1/2 𝑚 2 ,
90 degrees relative to the primary’s spin axis, then eccentric orbits thus 𝑣 𝑏1 ≈ 𝑣 𝑏𝑐1 cos 𝜃 is obtained by combining these two equations.
with a spin-orbit misalignment may be formed on the equatorial plane On the other hand, we have found that there is a negative correlation
of the secondary, and the spin-orbit angle is equal to this tilt angle. between the power-law exponent and the spin-orbit angle. Therefore,
For example, the formation of eccentric orbits during the generation for a spin-orbit misaligned system, its power law exponent is close to
of a massive radio pulsar binary or a high-mass X-ray binary from a the product of the power law exponent under conditions of spin-orbit
massive close binary through supernova explosion. alignment and the cosine of the spin-orbit angle.
The direction of the spin axis does not change in space in the
absence of external force, and any position of the primary on its
orbit can be regarded as a point on a simulated circular orbit that
5 CONCLUSIONS
takes the distance (represented by 𝑎) between the two components
as radius and the secondary as reference, so when the two spin axes • In a system with an eccentric orbit (e>0.01) and a mass ratio
are co-planar, the included angle (represented by 𝜃 1 ) between the order of magnitude greater than −7.5217 (excluding eccentric planets
line connecting the mass centers of the two components and the that have experienced planet-planet scattering), there is a power-law
equatorial plane (or a plane parallel to it) of the primary changes distribution relationship between the semi-major axis and the product
to 𝜃, and the angle (represented by 𝜃 2 ) between the direction of the of the mass ratio and the square of the rotation period ratio (see
rotational velocity of the primary’s gravitational field at the location equation (12)), and the exponent of this power-law distribution has
of the secondary and the direction of the reaction force acting on a positive linear relationship with the system’s mass ratio order of
the secondary (this reaction force is caused by the drag force acting magnitude and is greater than 0 and less than or equal to 1 (under
on the primary due to the rotation of the secondary’s gravitational conditions of spin-orbit alignment) (see equation (11), where both
field, the same below) changes to 0; when the plane passing through slope and intercept are divided by 3) and a negative correlation with
the primary’s spin axis and its proto spin axis is perpendicular to the system’s spin-orbit angle, and both of its components rotate super-
the plane passing through the two proto spin axes aligned with each synchronously (also including the case where its secondary rotates
other, 𝜃 1 changes to 0, and 𝜃 2 changes to 𝜃, as shown in figure 15. synchronously while the primary rotates super-synchronously and
The rotational velocity of the primary’s gravitational field spins down over time).
at the location of the secondary is (2𝜋𝑎 cos 𝜃 1 )/𝑃1 , that is, • There is no power law relationship between semi-major axis and
(2𝜋𝑎/𝑃1 ) cos 𝜃 1 , and the projection of this rotational velocity in the product of mass ratio and the square of rotation period ratio in
the direction of the reaction force acting on the secondary is equal to most systems with circular orbits (e<0.01); a system may have circu-
(2𝜋𝑎/𝑃1 ) cos 𝜃 1 cos 𝜃 2 , as shown in figure 15. As 𝜃 becomes small, lar orbits in one of the following circumstances: the semi-major axis
cos 𝜃 1 cos 𝜃 2 is closer or equal to cos 𝜃, because 𝑎 cos 𝜃 1 and 𝑎 cos 𝜃 2 of a system in which both components rotate super-synchronously
(which is close to or equal to the actual radius) are the two mutually (or only the secondary rotates synchronously while the primary ro-
perpendicular radii of such an ellipse, which is the projection of the tates super-synchronously) determined by equation (12) is smaller
primary’s simulated circular orbit on a plane which is the primary’s than its actual semi-major axis, a system with a mass ratio order
equatorial plane or a plane parallel to it and passing through the mass of magnitude is smaller than that of the Earth-moon system and in
center of the secondary, and their product is close to or equal to the which only the secondary rotates synchronously while the primary
product of the semi-minor axis (i.e. 𝑎 cos 𝜃) and the semi-major axis rotates super-synchronously, both components rotate synchronously,
(i.e. 𝑎 cos 0) of this ellipse, as shown in figure 15, so the projection the secondary rotates in a retrograde sense, and primary rotates syn-
of the rotational velocity of the primary’s gravitational field at the chronously or sub-synchronously; for a circular system, only when
location of the secondary in the direction of the reaction force act- its secondary rotates synchronously while the primary rotates super-
ing on the secondary is close to (2𝜋𝑎/𝑃1 ) cos 𝜃. Therefore, the drag synchronously and almost constantly over time, and its mass ratio
force acting on the secondary caused by the rotation of the primary’s order of magnitude is equal to that of the Earth-moon system or
gravitational field in the direction of the reaction force acting on the higher, there may be such a power law relationship in this system.
secondary is close to (2𝜋𝑎/𝑃1 ) cos 𝜃 (𝑚 1 /4𝜋𝑎 2 ) 1/2 𝑚 2 (see equation • Such a power law relationship played a decisive role in the
(17) as reference). formation of eccentric orbits.
Taking the primary as a reference, when 𝜃 equals to 0, the drag • Based on Einstein’s equivalence principle and Newton’s law of
force acting on the secondary caused by the rotation of the primary’s gravity, this study proposes that for a system in which both com-
gravitational field in the direction of the reaction force acting on the ponents rotate super-synchronously and two spin axes aligned with
secondary is represented by 𝑓𝑐1 , the circular orbit velocity caused by each other, the drag force acting on the primary in the rotating di-

MNRAS 000, 1–23 (20XX)

14 Yongfeng Dai (戴永丰)

θ When the plane passing through the primary’s spin axis and its
proto spin axis is perpendicular to the plane passing through the
two proto spin axes aligned with each other

The primary The primary’s simulated circu-

lar orbit, with the secondary as

0
os
reference

ac
θ a

θ!
90 deg
When the two spin a a cosθ! a When the two spin
axes are coplanar eg axes are coplanar
θ 90 d The secondary
θ)
v cos
sθ "(≈
sθ !co θ"
a co osθ

0
vc

os
ac
θ!
os
a vc

g s θ"
de co
90 ≈a
≈
e
it; th The secondary’s equatorial plane
l to e
ralle
la n e pa on of th
a p rojec ti
e or p
plan it is the is plane
rial n th
’s equato rawn) o rbit on
ary alf is d cular o
prim h ir
The e (only ulated c
s
ellip ry’s sim
a
prim

Figure 15. A schematic diagram of the projection (represented by a short arrow from the secondary) of the rotational velocity (represented by a dashed arrow
from the secondary) of the primary’s gravitational field at the location of the secondary in the direction of the reaction force (represented by a long arrow from
the secondary) acting on the secondary (this reaction force is caused by the drag force acting on the primary due to the rotation of the secondary’s gravitational
field, the same below). 𝜈 represents 2 𝜋𝑎/𝑃1 , 𝜃 is the angle between the two spin axes, 𝜃1 is the included angle between the line connecting the mass centers
of the two components and the equatorial plane (or a plane parallel to it) of the primary, 𝜃2 is the angle between the direction of the rotational velocity of the
primary’s gravitational field at the location of the secondary and the direction of the reaction force acting on the secondary, and 𝑎 is the distance between the
two components.

rection caused by the rotation of the secondary’s gravitational field

is 𝑚 1 (2𝜋𝑎/𝑃2 )(𝑚 2 /4𝜋𝑎 2 ) 1/2 , and the drag force acting on the sec-
ondary in the rotating direction caused by the rotation of the primary’s
gravitational field is 𝑚 2 (2𝜋𝑎/𝑃1 )(𝑚 1 /4𝜋𝑎 2 ) 1/2 (where 𝑚 1 and 𝑚 2
are the masses of the primary and the secondary respectively, 𝑃1 and
𝑃2 are the rotation periods of the primary and the secondary respec-
tively, and 𝑎 is the distance between the primary and the secondary).
In this system, the drag force acting on the secondary caused by the
rotation of the primary’s gravitational field results in a circular orbit
velocity (here we take one of the two components as reference). In
the presence of this circular orbit velocity, the orbital velocity caused
by the drag force acting on the primary due to the rotation of the
secondary’s gravitational field will be converted into the distance
between the two components (semi major axis), and this orbital ve-
locity is constrained by the ratio of the orbital radius of the primary
to that of the secondary (i.e. mass ratio), and this orbital velocity has
a power law relationship with the ratio of these two drag forces, and
this power law exponent is greater than 0 (e.g. the systems in which
the primary has an orbital radius of almost zero) to less than or equal
to 1 (e.g. the binaries whose two orbital radii are close to each other)
and is determined by the mass ratio.
• For a spin-orbit misaligned system, its power-law exponent is
close to the product of the power-law exponent under conditions of
spin-orbit alignment and the cosine of the spin-orbit angle.

MNRAS 000, 1–23 (20XX)

A power law relationship in an eccentric system, its role and the physical mechanism of its formation 15

Table 1. Orbital and rotation data for both components of eccentric systems.

𝑚 𝑃 𝑃𝑏
groups Name log(𝑚2 /𝑚1 ) 𝑒 log 𝑎 log(𝑟2 /𝑟1 )
(1022 kg) (h) (d)

i Sun A𝑎 198850000 ... 609.12 ... ... ... ...

Pallas B𝑎 0.0205 –9.9868 7.813 1683.80 0.2310 11.6177 –4.5902
Ceres B𝑎 0.0939 –9.3259 9.074 1680.15 0.0758 11.6171 –4.3265
Juno B𝑎 0.002 –10.9975 7.210 1592.49 0.2563 11.6013 –4.9503
Vesta B𝑎 0.0259 –9.8852 5.342 1325.86 0.0889 11.5482 –4.6674
Jupiter A𝑎 189819 ... 9.9250 ... ... ... ...
Elara B𝑎 0.00008 –9.3753 12 259.6528 0.217 10.0697 –3.0701
Himalia B𝑎 0.00095 –8.3006 9.6 250.5662 0.162 10.0592 –2.7765

ii Sun A𝑎 198850000 ... 609.12 ... ... ... ...

Mars B𝑎 64.171 –6.4912 24.6229 686.980 0.0935 11.3578 –3.0926
Earth B𝑎 597.23 –5.5224 23.9345 365.256 0.0167 11.1749 –2.7779
Mercury B𝑎 33.011 –6.7799 1407.6 87.969 0.2056 10.7628 –2.0174
Saturn A𝑎 56834 ... 10.656 ... ... ... ...
Iapetus B𝑎 0.181 –5.4969 S 79.33018 0.0283 9.5516 –0.3309

iii Sun A𝑎 198850000 ... 609.12 ... ... ... ...

Neptune B𝑎 10241.3 –4.2882 16.11 60189 0.0113 12.6527 –2.4811
Uranus B𝑎 8681.3 –4.3599 17.24* 30685.4 0.0457 12.4583 –2.4854
Saturn B𝑎 56834 –3.5439 10.656 10759.22 0.0565 12.1564 –2.3527
Jupiter B𝑎 189819 –3.0202 9.9250 4332.589 0.0489 11.8913 –2.1987
Saturn A𝑎 56834 ... 10.656 ... ... ... ...
Titan B𝑎 13.455 –3.6257 382.690104 15.94542 0.0292 9.0870 –0.1717

iv 61 Cygni A 139237000𝑏 ... 848.88𝑐 ... ... ... ...

61 Cygni B 125313000𝑏 –0.0458 908.16𝑐 263710𝑑 0.401𝑑 13.1219 0.004332
𝜉 Boo A 179018000𝑒 ... 148.8 𝑓 ... ... ... ...
𝜉 Boo B 131281000𝑒 –0.1347 276 𝑓 55337.2𝑔 0.512𝑔 12.6929 0.13397
𝛼 Cen A 219795000ℎ ... 540𝑖 ... ... ... ...
𝛼 Cen B 185782000ℎ –0.0730 1039.2 𝑗 29187.1ℎ 0.518ℎ 12.5465 0.16524
J0737-3039A 265943000 𝑘 ... 63.0528E-7 𝑘 ... ... ... ...
J0737-3039B 248637000 𝑘 –0.0292 77.0417E-5 𝑘 0.10225 𝑘 0.088 𝑘 8.9439 1.3816

Notes. Column 2: A indicates primary, B indicates secondary. Column 3: M⊙ takes 1.9891 × 1030 kg when mass is normalized. Column 4: mass ratio order of
magnitude. Column 5: * indicates sideways rotation; E indicates scientific notation. Column 6: yr takes 365.25 days when orbital period is normalized. Column
7: eccentricity. Column 8: logarithm of semi-major axis, where semi-major axis is measured in m; G takes 6.67384 × 10−11 Nm2 kg−2 when the semi-major axis
is calculated by Kepler’s third law using the orbital period. Column 9: logarithm of synchronous orbit radius ratio, obtained by equation (1). These values are
rounded to 0.0001, except those listed on Column 3.
References. (𝑎) Williams (2016), (𝑏) Henry (2010), (𝑐) Böhm-Vitense (2007), (𝑑) Josties (1983), (𝑒) Fernandes et al. (1998), ( 𝑓 ) Noyes et al. (1984), (𝑔) Wielen
(1962), (ℎ) Pourbaix et al. (2002), (𝑖) Bazot et al. (2007), ( 𝑗) Char et al. (1993) and (𝑘) Lyne et al. (2004).

MNRAS 000, 1–23 (20XX)

16 Yongfeng Dai (戴永丰)

Table 2. Orbital and rotation data for both components of eccentric stellar systems.

𝑚 𝑅 𝑣 sin 𝑖 𝑃 𝑃𝑏
groups Name log(𝑚2 /𝑚1 ) 𝑒 log 𝑎 log(𝑟2 /𝑟1 )
(M⊙ ) (R⊙ ) (km/s) (h) (d)

v 𝛼 Sco A 12.4𝑎 ... 883𝑎 20𝑏 <53608 ... ... ... ...
𝛼 Sco B 7.2𝑐 –0.2361 5.2𝑐 250𝑐 <25 935770𝑑 ... 13.8781 –2.2966
𝛼 And A 5.5𝑒 ... 2.7 𝑓 52 𝑓 <63.0 ... ... ... ...
𝛼 And B 2.3𝑒 –0.3786 1.65 𝑓 110 𝑓 <18.2 96.6963𝑒 0.60𝑒 11.0409 –0.4637
𝛼 CrB A 2.58𝑔 ... 3.04𝑔 139ℎ <26.5 ... ... ... ...
𝛼 CrB B 0.92𝑔 –0.4478 0.90𝑔 14𝑖 <78.1 17.36𝑔 0.370𝑔 10.4743 0.1629
NY Cep A 10.7 𝑗 6.0 𝑗 78 𝑗 <93.4
NY Cep B 8.8 𝑗 –0.0849 5.8 𝑗 155 𝑗 <45.4 15.276 𝑗 0.443 𝑗 10.6859 –0.2369
EP Cru A 5.02 𝑘 ... 3.590 𝑘 141.4 𝑘 <30.8 ... ... ... ...
EP Cru B 4.83 𝑘 –0.0167 3.495 𝑘 137.8 𝑘 <30.8 11.08 𝑘 0.1874 𝑘 10.4941 –0.0059
𝛿 Ori A 11.2𝑙 ... 13𝑙 157𝑙 <100 ... ... ... ...
𝛿 Ori B 5.6𝑙 –0.4559 5𝑙 138𝑙 <44 5.732𝑙 0.075𝑙 10.3805 –0.3396
𝛼 Vir A 10.25𝑚 ... 7.40𝑚 161𝑚 <55.8 ... ... ... ...
𝛼 Vir B 6.97𝑚 –0.1675 3.64𝑚 70𝑚 <63.1 4.015𝑚 0.067𝑚 10.2810 –0.0201

Notes. Column 2: A indicates primary, B indicates secondary. Column 3: mass. Column 4: mass ratio order of magnitude. Column 5: stellar radius. Column 6:
projected rotational velocity. Column 7: upper limit of rotation period, obtained by 𝑃 = 2 𝜋 𝑅/𝑣 sin 𝑖, where R⊙ takes 6.957 × 105 km. Column 8: orbital
period. Column 9: eccentricity. Column 10: logarithm of semi-major axis, where semi-major axis is measured in m; G takes 6.67384 × 10−11 Nm2 kg−2 and M⊙
takes 1.9891 × 1030 kg when the semi-major axis is calculated by Kepler’s third law using the orbital period. Column 11: logarithm of synchronous orbit radius
ratio, obtained by equation (2). These values are rounded to 0.0001.
References. (𝑎) Baade & Reimers (2007), (𝑏) Hoffleit & Warren (1991), (𝑐) Kudritzki & Reimers (1978), (𝑑) Reimers et al. (2008), (𝑒) Tomkin et al. (1995),
( 𝑓 ) Ryabchikovaet al. (1998), (𝑔) Tomkin & Popper (1986), (ℎ) Royer et al. (2002), (𝑖) Güdel et al (2003), ( 𝑗) Albrecht et al. (2011), (𝑘) Albrecht et al. (2013),
(𝑙) Harvin & Gies (2002) and (𝑚) Harrington et al. (2009).

Table 3. Orbital and rotation data for both components of spin-orbit misalignment systems with eccentric orbits.

𝑚 𝑃 𝑃𝑏 𝜃
groups Name log(𝑚2 /𝑚1 ) 𝑒 log 𝑎 log(𝑟2 /𝑟1 )
(M⊙ ) (s) (d) (deg)

vi LS 2883 20𝑎 ... 2 E+5 ... ... ... ... ...

B1259–63 1.4𝑎 –1.1549 0.0478𝑎 1236.7𝑎 0.8699𝑎 35𝑎 11.9715 –4.7994
B-star 20𝑏 ... 2 E+5𝑏 ... ... ... ... ...
J1740–3052 1.4𝑏 –1.1549 0.5703𝑏 231.03𝑏 0.5789𝑏 40–87𝑏 11.4858 –4.0816
B-star 8.8𝑐 ... 2.3 E+5𝑑 ... ... ... ... ...
J0045–7319 1.4𝑐 –0.7984 0.9263𝑐 51.169𝑐 0.808𝑐 25–41𝑐 10.9421 –3.8678
Earth A 3.0E-6𝑒 ... 8.6164E+4𝑒 ... ... ... ... ...
Moon B 3.7E-8𝑒 –1.9101 S𝑒 27.3217𝑒 0.0549𝑒 18.3–28.6𝑒 8.5848 0.3218

Notes. Column 3: mass; E indicates scientific notation. Column 4: mass ratio order of magnitude. Column 5: S indicates synchronous rotation; the rotation
period of the pulsar’s companion is an estimated value; E indicates scientific notation. Column 6: orbital period. Column 7: eccentricity. Column 8: spin-orbit
angle. Column 9: logarithm of semi-major axis, where semi-major axis is measured in m; G takes 6.67384 × 10−11 Nm2 kg−2 and M⊙ takes 1.9891 × 1030 kg
when the semi-major axis is calculated by Kepler’s third law using the orbital period. Column 10: logarithm of synchronous orbit radius ratio, obtained by
equation (1). These values are rounded to 0.0001.
References. (𝑎) Shannon et al. (2014), (𝑏) Madsen et al. (2012), (𝑐) Kaspi et al. (1996), (𝑑) Bell et al. (1995) and (𝑒) Williams (2016).

MNRAS 000, 1–23 (20XX)

A power law relationship in an eccentric system, its role and the physical mechanism of its formation 17

Table 4. Orbital and spin data for pulsars as secondaries in eccentric systems.

𝑃𝑏 𝑃 𝑚𝑐
Name 𝑒 1−𝑒
(d) (s) (M⊙ )

PSR J2032+41271 9131 0.95 0.05 0.1432 15

PSR J1638–47252 1940 0.95 0.2361 0.7639 Be
PSR B1259–633 1236.7 0.8699 0.1301 0.0478 20
4U 0352+304 250.3 0.111 0.889 837.67 17
GRO J1008–575 249.48 0.68 0.32 93.713 B0
GS 1843–026 242.18 0.88 0.12 94.473 9
PSR J1740–30527 231.03 0.5789 0.4211 0.5703 20
4U 1145–6198 186.5 >0.6 0.4 292 15
XTE J1946+2749 169.2 0.33 0.67 15.767 13
Swift J1626.6–515610 132.89 0.08 0.92 15.347 14
GX 304–111 132.19 0.524 0.476 275 10
A0535+26212 110.3 0.47 0.53 103.7 O9.7
GS 0834–43013 105.8 0.14 0.86 12.3 15
XTE J1543–56814 75.56 <0.03 0.97 27.122 10
PSR J0045–731915 51.169 0.8080 0.192 0.9263 10
EXO 2030+37516 46.021 0.41 0.59 41.472 B0
2S 1417–62417 42.12 0.4169 0.5831 17.545 >6
GX 301–218 41.498 0.462 0.538 680 55
KS 1947+30019 40.415 0.033 0.967 18.7 10
V 0332+5317 36.50 0.417 0.583 4.3763 Be
IGR J01054–725320 36.3 0.28 0.72 11.481 O9.5-B0
AX J0049–72921 33.38 0.40 0.6 74.867 8
2S 1553–54222 31.303 0.0351 0.9649 9.2788 Be
GRO J1750–2723 29.806 0.360 0.64 4.4535 Be
2E 0050.1–724721 28.51 0.41 0.59 8.9038 18
4U 0115+6317 24.317 0.342 0.658 3.6144 19
1A 1118–61624 24.0 <0.16 0.84 407.65 18
4U 1901+0325 22.583 0.0363 0.9637 2.7627 6
XTE J0103–72821 21.9 0.26 0.74 6.8508 20
SMC X–221 18.38 0.07 0.93 2.3719 22
XTE J0055–72726 17.79 0.43 0.57 18.385 B2
A0538–6627 16.65 >0.4 0.6 0.069 Be
IGR J00370+612228 15.663 0.48 0.52 346 22
1E 1145.1–614129 14.365 0.20 0.8 297.7 17
SAX J2103.5+454530 12.7 0.4 0.6 358.6 20
2S 0114+65031 11.59 0.16 0.84 9828 B1
OAO 1657–41532 10.448 0.107 0.893 38.22 14
1722–36333 9.7403 <0.19 0.81 413.9 22
4U 2206+5434,35 9.5591 0.15 0.85 5554 16
Vela X–136 8.9642 0.080 0.92 282.93 23
IGR J16320–475137,38 8.96 0.07 0.93 1303 Oe
4U1907+0939 8.3753 0.28 0.72 440.34 Be
IGR J16493–434840,41 6.782 <0.15 0.85 1093 47
XTE J1855–02642 6.0724 0.04 0.96 360.74 B0
IGR J18027–201643 4.5696 <0.2 0.8 139.61 24
X1908+07544 4.4007 0.021 0.979 604.68 20
IGR J16418–453245 3.7389 <0.2 0.8 1212 24
4U1538–5246 3.7284 0.174 0.826 528.81 20
IGR J16393–464347 3.6875 0.15 0.85 910.4 10

Notes. Column 2: orbital period. Column 3: eccentricity. Column 5: spin period. Column 6: companions’ mass or spectral type. These values are rounded to
0.0001.
References. (1) Lyne et al. (2015), (2) Lorimer et al. (2006), (3) Shannon et al. (2014), (4) Delgado-Martí et al. (2001), (5) Kühnel et al. (2013), (6) Finger et al.
(1999), (7) Madsen et al. (2012), (8) Cook & Warwick (1987), (9) Wilson et al. (2003), (10) Baykal et al. (2010), (11) Sugizaki et al. (2015), (12) Hill et al. (2007),
(13) Wilson et al. (1997), (14) In ’t Zand et al. (2001), (15) Kaspi et al. (1994), (16) Klochkov et al. (2007), (17) Raichur & Paul (2010), (18) Koh et al. (1997),
(19) Galloway et al. (2004), (20) Townsend et al. (2011b), (21) Townsend et al. (2011a), (22) Tsygankov et al. (2016), (23) Shaw et al. (2009), (24) Staubert et al.

(2011), (25) Galloway et al. (2005), (26) Schurch et al. (2009), (27) Skinner et al. (1982), (28) Grunhut et al. (2014), (29) Ray & Chakrabarty (2002), (30) Reig et al.
(2004), (31) Hall et al. (2000), (32) Mason et al. (2012), (33) Thompson et al. (2007), (34) Ribó et al. (2006), (35) Finger et al. (2010), (36) Nagase et al. (1984),
(37) Rodriguez et al. (2006), (38) Foschini L et al. (2004), (39) In ’t Zand et al. (1998), (40) Pearlman et al. (2019), (41) Cusumano et al. (2010), (42) Corbet &

Mukai (2002), (43) Hill et al. (2005), (44) Levine et al. (2004), (45) Sidoli et al. (2012), (46) Clark (2000) and (47) Thompson et al. (2006).

MNRAS 000, 1–23 (20XX)

18 Yongfeng Dai (戴永丰)

Table 5. Orbital and spin evolution data for pulsars as secondaries in eccentric systems.

investigation 𝑃𝑏 𝑃¤ 𝑏 𝑚𝑝 𝑚𝑐 𝑃 𝑃¤
Source name 𝑒
period (d) (10−9 ) (M⊙ ) (M⊙ ) (s) (10−10 )

J0045–7319𝑎 1991-1993 0.8080 51.17 ... 1.4 >4 0.926 446570

J0045–7319𝑏 1991-1994 0.8079 51 –∼279.3 1.4 8.8 ... ...
J1740–3052𝑐 1997-2011 0.5789 231.0 –∼3 1.4 20 0.570 0.0002
EXO 2030+375𝑑 1985-2006 ... ... ... ... ... 41.47 –162.75
EXO 2030+375𝑒 1985-2007 0.41 46.02 0 ... ... ... ...
2S 0114+650 𝑓 1996-2004 ... ... ... ... ... 9605 –8900
2S 0114+650𝑔 1996-1998 ... 11.591 0∼952 ... ... 9720 ...
2S 0114+650ℎ 1998-2000 0.18 11.598 0 1.4 16 ... ...
4U 1538–52𝑖 1997-2003 0.18 3.728 0∼ ... ... 526.8 –50∼–390
OAO 1657–415 𝑗 1978-2005 ... ... ... ... ... 37 –12.4
OAO 1657–415 𝑗 1991-2005 0.104 10.448 0 ... ... ... ...
OAO 1657–415 𝑘 1991-2008 0.1075 10.447 –97.4 1.4 16 37 –∼0.3
Vela X–1𝑙 1975-1982 0.080 8.964 0 ... ... 282.93 –∼0

Notes. Column 3: eccentricity. Column 5: orbital period derivative; 0∼means that an increase in orbital period was observed over the whole investigation
period. Column 6: pulsar’s mass. Column 7: companion’s mass. Column 9: pulsar’s spin period derivative. These values are rounded to 0.0001.
References. (𝑎) Kaspi et al. (1994), (𝑏) Kaspi et al. (1996), (𝑐) Madsen et al. (2012), (𝑑) Klochkov et al. (2007), (𝑒) Wilson et al. (2008), ( 𝑓 ) Bonning & Falanga
(2005), (𝑔) Corbet et al. (1999), (ℎ) Grundstrom et al. (2007), (𝑖) Mukherjee et al. (2006), ( 𝑗) Barnstedt et al. (2008), (𝑘) Jenke et al. (2012), (𝑙) Nagase et al.
(1984).

Table 6. Orbital and spin data for the first-born pulsars in eccentric double neutron star binaries.

𝑃𝑏 𝑃
Name 𝑒 1−𝑒
(d) (ms)

J0453+1559 4.072 0.113 0.887 45.8

J0737–3039 0.102 0.088 0.912 22.7
J1518+4904 8.634 0.249 0.751 40.9
B1534+12 0.421 0.274 0.726 37.9
J1753–2240 13.638 0.304 0.696 95.1
J1756–2251 0.320 0.181 0.819 28.5
J1811–1736 18.779 0.828 0.172 104.2
J1829+2456 1.176 0.139 0.861 41.0
J1913+1102 0.206 0.090 0.91 27.3
B1913+16 0.323 0.617 0.383 59.0
J1930–1852 45.060 0.399 0.601 185.5

Notes. Column 2: orbital period. Column 3: eccentricity. Column 5: spin period.

References. Tauris et al. (2017).

Table 7. Orbital and spin evolution data for pulsars in eccentric double neutron star binaries.

𝑃𝑏 𝑃¤ 𝑏 𝑚𝑝 𝑚𝑐 𝑃 𝑃¤
Name 𝑒
(d) (10−12 ) (M⊙ ) (M⊙ ) (ms) (10−20 )

B2127+11C𝑎 0.6814 0.3353 –3.96 1.358 1.354 30.53 498.79

B1913+16𝑏 0.6171 0.3230 –2.427 1.442 1.386 59.03 862.71
B1534+12𝑐 0.2737 0.4207 –0.137 1.333 1.345 37.90 242.26
J1518+4904𝑑 0.2495 8.6340 –0.24 0.72 2.00 40.93 2.72
J1756–2251𝑒 0.1806 0.3196 –0.229 1.341 1.230 28.46 101.75
J1141–6545 𝑓 0.1719 0.1976 –0.43 1.30 0.986 393.9 429459
J0737–3039A𝑔 0.0878 0.1022 –1.252 1.338 1.249 22.70 175.99
J0737–3039B𝑔 0.0878 0.1022 –1.252 1.249 1.338 2773 89228
J1906+0746ℎ 0.0853 0.1660 –0.56 1.291 1.322 144.1 2026783

Notes. Column 2: eccentricity. Column 4: orbital period derivative. Column 5: pulsar’s mass. Column 6: companion’s mass. Column 8: pulsar’s spin period
derivative. These values are rounded to 0.0001.
References. (𝑎) Jacoby et al. (2006), (𝑏) Taylor & Weisberg (1989), (𝑐) Stairs et al. (2002), (𝑑) Janssen et al. (2008), (𝑒) Ferdman et al. (2014), ( 𝑓 ) Bailes et al.
(2003), (𝑔) Kramer et al. (2006); (ℎ) Van Leeuwen et al. (2015).

MNRAS 000, 1–23 (20XX)

A power law relationship in an eccentric system, its role and the physical mechanism of its formation 19

Table 8. Orbital and rotation data for both components of circular systems.

𝑚 𝑃 𝑃𝑏
groups Name log(𝑚2 /𝑚1 ) 𝑒 log 𝑎 log(𝑟2 /𝑟1 )
(1022 kg) (h) (d)

a Jupiter A𝑎 189819 ... 9.9250 ... ... ... ...

Amalthea B𝑎 0.00075 –8.4033 S 0.498179 0.003 8.2586 –2.7472
Metis B𝑎 0.00001 –10.2783 S 0.294779 0.0002 8.1072 –3.5241
Saturn A𝑎 56834 ... 10.656 ... ... ... ...
Janus B𝑎 0.000190 –8.4758 S 0.6945 0.007 8.1803 –2.6957
EpimetheusB𝑎 0.000053 –9.0303 S 0.6942 0.009 8.1802 –2.8807
Mars A𝑎 64.171 ... 24.6229 ... ... ... ...
Deimos B𝑎 0.00000024 –8.4271 S 1.26244 0.0005 7.3703 –2.7490

b Saturn A𝑎 56834 ... 10.656 ... ... ... ...

Rhea B𝑎 0.231 –5.3910 S 4.517500 0.0010 8.7218 –1.1253
Dione B𝑎 0.11 –5.7132 S 2.736915 0.0022 8.5768 –1.3778
Tethys B𝑎 0.0618 –5.9636 S 1.887802 0.0000 8.4693 –1.5688
EnceladusB𝑎 0.0108 –6.7212 S 1.370218 0.0045 8.3766 –1.9141
Uranus A𝑎 8681.3 ... 17.24 ... ... ... ...
Miranda B𝑎 0.0066 –6.1190 S 1.413479 0.0013 8.1136 –1.8437

c Jupiter A𝑎 189819 ... 9.9250 ... ... ... ...

Callisto B𝑎 10.759 –4.2466 S 16.68902 0.007 9.2748 –0.3449
Ganymede B𝑎 14.819 –4.1075 S 7.154553 0.001 9.0295 –0.5438
Europa B𝑎 4.800 –4.5971 S 3.551181 0.009 8.8268 –0.9098
Io B𝑎 8.932 –4.3274 S 1.769138 0.004 8.6251 –1.0216
Uranus A𝑎 8681.3 ... 17.24 ... ... ... ...
Oberon B𝑎 0.301 –4.4600 S 13.46324 0.0008 8.7661 –0.6381
Titania B 𝑎 0.352 –4.3920 S 8.705872 0.0022 8.6394 –0.7417
Umbriel B𝑎 0.117 –4.8704 S 4.144177 0.0050 8.4254 –1.1161
Ariel B𝑎 0.135 –4.8083 S 2.520379 0.0034 8.2811 –1.2393

Sun A𝑎 198850000 ... 609.12 ... ... ... ...

Venus B𝑎 486.75 –5.6112 -5832.6 224.701 0.0067 11.0343 -1.2163

Pluto A𝑎 1.303 ... S ... ... ... ...

Charon B𝑎 0.1586 –0.9146 S 6.3872 0.0 7.2922 -0.3049

𝛼 Aur A 509209000𝑏 ... 207.36𝑐 ... ... ... ...

𝛼 Aur B 535068000𝑏 0.0215 2544𝑐 104𝑏 0.000𝑏 11.0513 0.73304

Notes. Column 2: A indicates primary, B indicates secondary. Column 3: M⊙ takes 1.9891 × 1030 kg when mass is normalized. Column 4: mass ratio order of
magnitude. Column 5: S indicates synchronous rotation, – indicates rotating in a retrograde sense. Column 6: orbital period. Column 7: eccentricity. Column 8:
logarithm of semi-major axis, where semi-major axis is measured in m; G takes 6.67384 × 10−11 Nm2 kg−2 when the semi-major axis is calculated by Kepler’s
third law using the orbital period. Column 9: logarithm of synchronous orbit radius ratio, obtained by equation (1). These values are rounded to 0.0001, except
those listed on Column 3.
References. (𝑎) Williams (2016), (𝑏) Hummel et al. (1994) and (𝑐) Strassmeier et al. (2001).

Table 9. Orbital and rotation data for both components of circular stellar systems.

𝑚 𝑅 sin 𝑖 𝑃 𝑃𝑏
Name log(𝑚2 /𝑚1 ) 𝑒 log 𝑎 log(𝑟2 /𝑟1 )
(M⊙ ) (R⊙ ) (km/s) (h) (d)

𝛽 Aur A 2.327𝑎 ... 2.63𝑏 34𝑏 <94 ... ... ... ...
𝛽 Aur B 2.389𝑎 0.011 2.77𝑏 33𝑏 <102 3.960𝑎 0.0𝑎 10.089 –0.020

Notes. Column 1: A indicates primary, B indicates secondary. Column 2: mass. Column 3: mass ratio order of magnitude. Column 4: stellar radius. Column 5:
projected rotational velocity. Column 6: upper limit of rotation period, obtained by 𝑃 = 2 𝜋 𝑅/𝑣 sin 𝑖, where R⊙ takes 6.957 × 105 km. Column 7: orbital
period. Column 8: eccentricity. Column 9: logarithm of semi-major axis, where semi-major axis is measured in m; G takes 6.67384 × 10−11 Nm2 kg−2 and M⊙
takes 1.9891 × 1030 kg when the semi-major axis is calculated by Kepler’s third law using the orbital period. Column 10: logarithm of synchronous orbit radius
ratio, obtained by equation (2).
References. 𝑎 Behr et al. (2011) and 𝑏 Nordstrom & Johansen (1994)

MNRAS 000, 1–23 (20XX)

20 Yongfeng Dai (戴永丰)

Table 10. Orbital and spin evolution data for pulsars as secondaries in circular systems.

investigation 𝑃𝑏 𝑃¤ 𝑏 𝑚𝑝 𝑚𝑐 𝑃 𝑃¤
Source name 𝑒
period (d) (10−9 ) (M⊙ ) (M⊙ ) (s) (10−10 )

LMC X–4𝑎 1977-1991 0.006 1.408 –2.04 1.38 14.7 13.5 –∼0
Cen X–3𝑏 1971-1981 0.0008 2.087 –10.3 1.0 19 ... ...
Cen X–3𝑐 1971-1981 ... 2.087 ... ... ... 4.8317 –0.15∼–0.46
SMC X–1𝑑 1971-2003 0.0009 3.892 –36.25 1.06 17.2 0.7074 –0.18

Notes. Column 3: eccentricity. Column 5: orbital period derivative. Column 6: pulsar’s mass. Column 7: companion’s mass. Column 9: pulsar’s spin period
derivative. These values are rounded to 0.0001.
References. (𝑎) Woo et al. (1996), (𝑏) Kelley et al. (1983), (𝑐) Murakami et al. (1983) and (𝑑) Inam et al. (2010).

Table 11. Orbital and spin evolution data for pulsars as primaries in circular systems.

𝑃𝑏 𝑃¤ 𝑏 𝑚𝑝 𝑚𝑐 𝑃 𝑃¤
Name 𝑒
(d) (10−12 ) (M⊙ ) (M⊙ ) (ms) (10−20 )

J2019+2425𝑎 1.11 E-4 76.512 –(< 90) 1.33 0.35 3.934 0.7024
J2051–0827𝑏 6.2 E-5 0.0991 –18.1 1.8 0.05 4.508 1.2733
J1701–3006B𝑐 <4.5 E-5 0.1445 –5.12 1.4 >0.12 3.593 –34.83
J2145–0750𝑑 1.93 E-5 6.8389 –(<2.4) 1.4 0.51 16.05 2.9757
J0437–4715𝑒 1.92 E-5 5.7410 –3.73 1.76 0.254 5.757 5.7294
J0348+0432 𝑓 2 E-6 0.1024 –0.273 2.01 0.172 39.12 24.073
J1012+5307𝑔 <1.3 E-6 0.6047 –(<0.1) 1.68 0.16 5.256 1.7134
J0751+1807ℎ 5 E-7 0.2631 –0.064 2.1 0.191 3.479 0.7786
J1738+0333𝑖 3.4 E-7 0.3548 –0.026 1.46 0.181 5.85 2.4120
B1957+20 𝑗 <4 E-5 0.3820 14.7 1.4 0.025 1.607 1.68515
4U 1822–37 𝑘 ... 0.2321 130 1.61 0.44 593 –2.48E+8
SAX J1808.4–3658𝑙,𝑚 <1.2E-4 0.0839 3.80 2.2 0.07 2.494 0.3421

Notes. Column 2: eccentricity; E indicates scientific notation. Column 4: orbital period derivative. Column 5: pulsar’s mass. Column 6: companion’s mass.
Column 8: pulsar’s spin period derivative; E indicates scientific notation. These values are rounded to 0.0001.
References. (𝑎) Nice et al. (2001), (𝑏) Lazaridis et al. (2011), (𝑐) Lynch et al. (2012), (𝑑) Löhmer et al. (2004), (𝑒) Verbiest et al. (2008); ( 𝑓 ) Antoniadis et al.
(2013), (𝑔) Lange et al. (2001), (ℎ) Nice et al. (2005), (𝑖) Freire et al. (2012), ( 𝑗) Ryba & Taylor (1991), (𝑘) Jain et al. (2010), (𝑙) Hartman et al. (2009) and
(𝑚) Deloye et al. (2008).

MNRAS 000, 1–23 (20XX)

A power law relationship in an eccentric system, its role and the physical mechanism of its formation 21
DATA AVAILABILITY Fernandes J., Lebreton Y., Baglin A., Morel, P. 1998, A&A, 338, 455. https:
//adsabs.harvard.edu/pdf/1998A%26A...338..455F
Full citations that include URLs or DOIs are provided for all raw Finger M. H., Bildsten L., Chakrabarty D., Prince T. A., Scott D. M., Wilson C.
data used in this study (see tables 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 and 11). A., Wilson R. B., Zhang S. N. 1999, ApJ, 517, 449, doi:10.1086/307191
Finger M. H., Ikhsanov N. R., Wilson-Hodge C. A., Patel S. K. 2010, ApJ,
709, 1249, doi:10.1088/0004-637X/709/2/1249
Foschini L., Tomsick J. A., Rodriguez J., Walter R., Goldwurm A., Corbel
REFERENCES S., Kaaret P. High-energy emission from IGR J16320-4751. The IN-
Albrecht S., Setiawan J., Torres G., Fabrycky D. C., Winn J. N. 2013, ApJ, TEGRAL Universe. Proceedings of the Fifth INTEGRAL Workshop.
767, 32, doi:10.1088/0004-637X/767/1/32 16-20 February 2004, Munich, Germany. Editor: B. Battrick, Scientific
Albrecht S., Winn J. N. Carter J. A., Snellen I. A. G., de Mooij E. J. W. 2011, Editors: V. Schoenfelder, G. Lichti, & C. Winkler. ESA SP-552, Noord-
ApJ, 726, 68, doi:10.1088/0004-637X/726/2/68 wijk: ESA Publication Division, ISBN 92-9092-863-8, 2004, p. 247-252.
Albrecht S., et al. 2012, ApJ, 757, 18, doi:10.1088/0004-637X/757/1/18 http://arxiv.org/abs/astro-ph/0403479
Antoniadis J., et al. 2013, Science, 340, 448, doi:10.1126/science.1233232 Freire P. C. C., et al. 2011, MNRAS, 412, 2763, doi:10.1111/j.1365-
Ashby N. 2004, Nature, 431, 918, doi:10.1038/431918a 2966.2010.18109.x
Baade R., Reimers, D. 2007, A&A, 474, 229, doi:10.1051/0004-6361:200 Freire P. C. C., et al. 2012, MNRAS, 423, 3328, doi:10.1111/j.1365-
77308 2966.2012.21253.x
Bailes M., Ord S. M., Knight H. S., Hotan A. W. 2003, ApJ, 595, L49, Galloway D. K., Morgan E. H., Levine A. M. 2004, ApJ, 613, 1164,
doi:10.1086/378939 doi:10.1086/423265
Barnstedt J., et al. 2008, A&A, 486, 293, doi:10.1051/0004-6361:20078707 Galloway D. K., Wang Z., Morgan E. H. 2005, ApJ, 635, 1217,
Baykal A., Gögüs E., Inam S. C., Belloni T. 2010, ApJ, 711, 1306, doi:10.1086/497573
doi:10.1088/0004-637X/711/2/1306 Grundstrom E. D., et al. 2007, ApJ, 656, 431, doi:10.1086/510508
Bazot M., Bouchy F., Kjeldsen H., Charpinet S., Laymand M., Vauclair S. Grunhut J. H., Bolton C. T., McSwain M. V. 2014, A&A, 563, A1,
2007, A&A, 470, 295, doi:10.1051/0004-6361:20065694 doi:10.1051/0004-6361/201322738
Behr B. B., Cenko A. T., Hajian A. R., McMillan R. S., Murison M., Meade Güdel M., Arzner K., Audard M., Mewe R. 2003, A&A, 403, 155,
J., Hindsley R. 2011, AJ, 142, 6, doi:10.1088/0004-6256/142/1/6 doi:10.1051/0004-6361:20030257
Bell J. F., Bessell M. S., Stappers B. W., Bailes M., Kaspi V. M. 1995, ApJ, Hall T. A., Finley J. P., Corbet R. H. D., Thomas R. C. 2000, ApJ, 536, 450,
447, L117, doi:10.1086/309565 doi:10.1086/308924
Böhm-Vitense E. 2007, ApJ, 657, 486, doi:10.1086/ 510482 Harrington D., Koenigsberger G., Moreno E., Kuhn J. 2009, ApJ, 704, 813,
Bonning E. W., Falanga M. 2005, A&A, 436, L31, doi:10.1051/0004- doi:10.1088/0004-637X/704/1/813
6361:200500117 Hartman J. M., Patruno A., Chakrabarty D., Markwardt C. B., Morgan E. H.,
Bowler B. P., Blunt S. C., Nielsen E. L. 2020, AJ, 159, 63 doi:10.3847/1538- van der Klis M., Wijnands R. 2009, ApJ, 702, 1673, doi:10.1088/0004-
3881/ab5b11 637X/702/2/1673
Brewer J. M., Fischer D. A., Madhusudhan N. 2017, AJ, 153, 83, Harvin J. A., Gies D. R. 2002, International Astronomical Union Colloquium,
doi:10.3847/1538-3881/153/2/83 187, 47, doi:10.1017/S0252921100001214
Bryan M. L., et al. 2016, ApJ, 821, 89, doi:10.3847/0004-637X/821/2/89 Henry T. J. 2010, the 100 Nearest Star Systems, RECONS, Univ. Georgia
Campbell L., Moffat J. 1983, ApJ, 275, L77, doi:10.1086/184174 State. http://www.solstation.com/stars/61cygni2.htm
Chapront J., Chapront-Touzé M., Francou G. 2002, A&A, 387, 700, Hill A. B., et al. 2007, MNRAS, 381, 1275, doi:10.1111/j.1365-
doi:10.1051/0004-6361:20020420 2966.2007.12326.x
Char S., Foing B. H., Beckman J., Garcia Lopez R. J., Rebolo R. 1993, A&A, Hill A. B., et al. 2005, A&A, 439, 255, doi:10.1051/0004-6361:20052937
276, 78. http://adsabs.harvard.edu/pdf/1993A%26A...276... Hoffleit D., Warren W. H. Jr. 1991, HR 6134, database en-
78C try, The Bright Star Catalogue, 5th Revised Ed. http:
Clark G. W. 2000, ApJ, 542, L131, doi:10.1086/312926 //webviz.u-strasbg.fr/viz-bin/VizieR-5?-out.add=.&
Cook M. C., Warwick R. S. 1987, MNRAS, 225, 369, -source=V/50/catalog&recno=6134
doi:10.1093/mnras/225.2.369 Howe K. S., Clarke C. J. 2009, MNRAS, 392, 448, doi:10.1111/j.1365-
Corbet R. H. D. 1984, A&A, 141, 91. https://ui.adsabs.harvard.edu/ 2966.2008.14073.x
abs/1984A&A...141...91C Hummel C. A., Armstrong J. T., Quirrenbach A., Buscher D. F., Mozurkewich
Corbet R. H. D., Finley J. P., Peele A. G. 1999, ApJ, 511, 876, D., Elias II N. M. 1994, AJ, 107, 1859, doi:10.1086/116995
doi:10.1086/306727 Inam S. C., Baykal A., Beklen E. 2010, MNRAS, 403, 378,
Corbet R. H. D., Mukai K. 2002, ApJ, 577, 923, doi:10.1086/342244 doi:10.1111/j.1365-2966.2009.16121.x
Cossou C., Raymond S. N., Hersant F., Pierens A. 2014, A&A 569, A56, In ’t Zand J. J. M., Baykal A., Strohmayer T. E. 1998, ApJ, 496, 386,
doi:10.1051/0004-6361/201424157 doi:10.1086/305362
Cridland A. J., van Dishoeck E. F., Alessi M., Pudritz R. E. 2019, A&A, 632, In ’t Zand J. J. M., Corbet R. H. D., Marshall F. E. 2001, ApJ, 553, L165,
A63, doi:10.1051/0004-6361/201936105 doi:10.1086/320688
Cusumano G., La Parola V., Romano P., Segreto A., Vercellone S., Chincarini Jacoby B. A., Cameron P. B., Jenet F. A., Anderson S. B., Murty R. N.,
G. 2010, MNRAS, 406, L16, doi:10.1111/j.1745-3933.2010.00872.x Kulkarni S. R. 2006, ApJ, 644, L113, doi:10.1086/505742
Dawson R. I., Murray-Clay R. A. 2013, ApJL, 767, L24, doi:10.1088/2041- Jain C., Paul B., Dutta A. 2010, MNRAS, 409, 755, doi:10.1111/j.1365-
8205/767/2/L24 2966.2010.17336.x
Delgado-Martí H., Levine A. M., Pfahl E., Rappaport S. A. 2001, ApJ, 546, Janssen G. H., Stappers B. W., Kramer M., Nice D. J., Jessner A., Cognard I.,
455, doi:10.1086/318236 Purver M. B. 2008, A&A, 490, 753, doi:10.1051/0004-6361:200810076
Deloye C. J., Heinke C. O., Taam R. E., Jonker P. G. 2008, MNRAS, 391, Jenke P. A., Finger M. H., Wilson-Hodge C. A., Camero-Arranz A. 2012,
1619, doi:10.1111/j.1365-2966.2008.14021.x ApJ, 759, 124, doi:10.1088/0004-637X/759/2/124
Dewi J. D. M., Podsiadlowski Ph., Pols O. R. 2005, MNRAS, 363, L71, Josties F. J. 1983, International Astronomical Union Colloquium, 62, 16,
doi:10.1111/j.1745-3933.2005.00085.x doi:10.1017/S0252921100009763
Faucher-Giguère C., Kaspi V. M. 2006, ApJ, 643, 332, doi:/10.1086/501516 Kane S. R., Ciardi D. R., Gelino D. M., von Braun K. 2012, MNRAS, 425,
Faulkner A. J., et al. 2005, ApJ, 618, L119, doi:10.1086/427776 757, doi:10.3847/1538-3881/ab04a9
Ferdman R. D., et al. 2013, ApJ, 767, 85, doi:10.1088/0004-637X/767/1/85 Kaspi V. M., Bailes M., Manchester R. N., Stappers B. W., Bell J. F., 1996,
Ferdman R. D., et al. 2014, MNRAS, 443, 2183, doi:/10.1093/ mnras/stu1223 Nature, 381, 584, doi:10.1038/381584a0

MNRAS 000, 1–23 (20XX)

22 Yongfeng Dai (戴永丰)
Kaspi V. M., Johnston S., Bell, J. F., Manchester R.N., Bailes M., Bessell M., 897, doi:10.1051/0004-6361:20020943
Lyne A.G., D’Amico N. 1994, ApJ, 423, L43, doi:10.1086/187231 Ryabchikova T., Malanushenko V., Adelman, S. J. 1998, Contrib. Astron.
Kawaler S. D. 2004, Symposium-International Astronomical Union, 215, Obs. Skalnate Pleso, 27, 356. http://adsabs.harvard.edu/pdf/
561, doi:10.1017/S0074180900196196 1998CoSka..27..356R
Kelley R. L., Rappaport S., Clark G. W., Petro L. D. 1983, ApJ, 268, 790, Ryba M. F., Taylor J. H. 1991, ApJ, 380, 557. http://adsabs.harvard.
doi:10.1086/161001 edu/pdf/1991ApJ...380..557R
Klochkov D., et al. 2007, A&A, 464, L45, doi:10.1051/0004-6361:20066801 Schurch M. P. E., et al. 2009, MNRAS, 392, 361, doi:10.1111/j.1365-
Koh D. T., et al. 1997, ApJ, 479, 933, doi:10.1086/303929 2966.2008.14048.x
Kramer M., et al. 2006, Science, 314, 97, doi:10.1126/science.1132305 Shannon R. M., Johnston S., Manchester R. N. 2014, MNRAS, 437, 3255,
Kudritzki R. P., Reimers D. 1978, A&A, 70, 227. http://adsabs. doi:10.1093/mnras/stt2123
harvard.edu/pdf/1978A%26A....70..227K Shaw S. E., Hill A. B., Kuulkers E., Brandt S., Chenevez J., Kretschmar P.
Kühnel M., et al. 2013, A&A, 555, A95, doi:10.1051/0004-6361/201321203 2009, MNRAS, 393, 419, doi:10.1111/j.1365-2966.2008.14212.x
Lange C., Camilo F., Wex N., Kramer M., Backer D. C., Lyne A. Sidoli L., Mereghetti S., Sguera V., Pizzolato F. 2012, MNRAS, 420, 554,
G., Doroshenko O. 2001, MNRAS, 326, 274, doi:10.1046/j.1365- doi:10.1111/j.1365-2966.2011.20063.x
8711.2001.04606.x Skinner G. K., Bedford D. K., Elsner R. F., Leahy D., Weisskopf M. C.,
Lazaridis K., et al. 2011, MNRAS, 414, 3134, doi:10.1111/j.1365- Grindlay J. 1982, Nature, 297, 568, doi:10.1038/297568a0
2966.2011.18610.x Stairs H., Thorsett S. E., Taylor J. H., Wolszczan A. 2002, ApJ, 581, 501,
Levine A. M., Rappaport S., Remillard R., Savcheva A. 2004, ApJ, 617, 1284, doi:10.1086/344157
doi:10.1086/425567 Staubert R., Pottschmidt K., Doroshenko V., Wilms J., Suchy S., Roth-
Li X. D., van den Heuvel E. P. J. 1996, A&A, 314, L13. http://adsabs. schild R., Santangelo A. 2011, A&A, 527, A7, doi:10.1051/0004-
harvard.edu/pdf/1996A%26A...314L..13L 6361/201015737
Löhmer O., Kramer M., Driebe T., Jessner A., Mitra D., Lyne A. G. 2004, Stiles B. W., et al. 2008, AJ, 135, 1669, doi:10.1088/0004-6256/135/5/1669
A&A, 426, 631, doi:10.1051/0004-6361:20041031 Strassmeier K. G., Reegen P., Granzer T. 2001, AN, 322,
Lorimer D. R., et al. 2006, MNRAS, 372, 777, doi:10.1111/j.1365- 115, doi:10.1002/1521-3994(200106)322:2<115::AID-ASN
2966.2006.10887.x A115>3.0.CO:2-J
Lynch R. S., Freire P. C. C., Ransom S. M., Jacoby B. A. 2012, ApJ, 745, Sugizaki M., Yamamoto T., Mihara T., Nakajima M., Makishima K. 2015,
109, doi:10.1088/0004-637X/745/2/109 PASJ, 67, 73-1, doi:10.1093/pasj/psv039
Lyne A. G., et al. 2004, Science, 303, 1153, doi:10.1126/science.1094645 Tauris T. M., et al. 2017, ApJ, 846, 170, doi:10.3847/1538-4357/aa7e89
Lyne A. G., Stappers B. W., Keith M. J., Ray P. S., Kerr M., Camilo F., Taylor J. H., Weisberg J. M. 1989, ApJ, 345, 434, doi:10.1086/167917
Johnson T. J. 2015, MNRAS, 451, 581, doi:10.1093/mnras/stv236 Thompson T. W. J., Tomsick J. A., in ’t Zand J. J. M., Rothschild R. E., Walter
Madsen E. C., et al. 2012, MNRAS, 425, 2378, doi:10.1111/j.1365- R. 2007, ApJ, 661, 447, doi:10.1086/513458
2966.2012.21691.x Thompson T. W. J., Tomsick J. A., Rothschild R. E., in ’t Zand J. J. M., Walter
Mason A. B., Clark J. S., Norton A. J., Crowther P. A., Tauris T. M., Langer N., R. 2006, ApJ, 649, 373, doi:10.1086/506251
Negueruela I., Roche P. 2012, MNRAS, 422, 199, doi:10.1111/j.1365- Tomkin J., Popper D. M. 1986, AJ, 91, 1428, doi:10.1086/114121
2966.2012.20596.x Tomkin J., Pan X., McCarthy J. K. 1995, AJ, 109, 780, doi:10.1086/117321
McQuillan A., Mazeh T., Aigrain S. 2013, ApJ, 775, L11, doi:10.1088/2041- Torres G., Claret A., Young P. A. 2009, ApJ, 700, 1349, doi:10.1088/0004-
8205/775/1/L11 637X/700/2/1349
Mukherjee U., Raichur H., Paul B., Naik S., Bhatt N. 2006, JApA, 27, 411, Townsend L. J., Coe M. J., Corbet R. H. D., Hill A. B. 2011a, MNRAS, 416,
doi:10.1007/BF02709367 1556, doi:10.1111/j.1365-2966.2011.19153.x
Muley D., Fung J., van der Marel N. 2019, ApJ, 879, L2, doi:10.3847/2041- Townsend L. J., et al. 2011b, MNRAS, 410, 1813, doi:10.1111/j.1365-
8213/ab24d0 2966.2010.17563.x
Murakami T., et al. 1983, ApJ, 264, 563. http://adsabs.harvard.edu/ Triaud A. H. M. J., et al. 2010, A&A, 524, A25, doi:10.1051/0004-
pdf/1983ApJ...264..563M 6361/201014525
Nagase F., Hayakawa S., Kunieda H., Masai K., Sato N., Tawara Y. 1984, Tsygankov S. S., Lutovinov A. A., Krivonos R. A., Molkov S. V.,
ApJ, 280, 259, doi:10.1086/161993 Jenke P. J., Finger M. H., Poutanen J. 2016, MNRAS, 457, 258,
Nice D. J., Splaver E. M., Stairs I. H. 2001, ApJ, 549, 516, doi:10.1086/319079 doi:10.1093/mnras/stv2849
Nice D. J., Splaver E. M., Stairs I. H., Löhmer O., Jessner A., Kramer M., Van Leeuwen J., et al. 2015, ApJ, 798, 118, doi:10.1088/0004-
Cordes J. M. 2005, ApJ, 634, 1242, doi:10.1086/497109 637X/798/2/118
Nordstrom B., Johansen, K. T. 1994, A&A, 291, 777. http://adsabs. Verbiest J. P. W., et al. 2008, ApJ, 679, 675, doi:10.1086/529576
harvard.edu/pdf/1994A%26A...291..777N Walkowicz L. M., Basri G. S. 2013, MNRAS, 436, 1883,
Noyes R. W., Hartmann L. W., Baliunas S. L., Duncan D. K., Vaughan A. H. doi:10.1093/mnras/stt1700
1984, ApJ, 279, 763, doi:10.1086/161945 Weiss L. M., et al. 2018, AJ, 156, 254, doi:10.3847/1538-3881/aae70a
Pearlman A. B., Coley J. B., Corbet R. H. D., Pottschmidt K. 2019, ApJ, 873, Wielen R. 1962, AJ, 67, 599, doi:10.1086/108791
86, doi:10.3847/1538-4357/aaf001 Williams D. R. 2016, Planetary Fact Sheets, NASA, Last accessed 2016.
Portegies Zwart S. F. 2000, ApJ, 544, 437, doi:10.1086/317190 http://nssdc.gsfc.nasa.gov/planetary/factsheet/
Pourbaix D., et al. 2002, A&A 386, 280, doi:10.1051/0004-6361:20020287 Williams G. E. 2000, Reviews of Geophysics, 38, 37,
Raichur H., Paul B. 2010, MNRAS, 406, 2663, doi:10.1111/j.1365- doi:10.1029/1999RG900016
2966.2010.16862.x Wilson C. A., Finger M. H., Camero-Arranz A. 2008, ApJ, 678, 1263,
Ray P. S., Chakrabarty D. 2002, ApJ, 581, 1293, doi:10.1086/344300 doi:10.1086/587134
Reig P., Negueruela I., Fabregat J., Chato R., Blay, P., Mavromatakis F. 2004, Wilson C. A., Finger M. H., Coe M. J., Negueruela I. 2003, ApJ, 584, 996,
A&A, 421, 673, doi:10.1051/0004-6361:20035786 doi:10.1086/345791
Reimers D., Hagen H.-J., Baade R., Braun K. 2008, A&A, 491, 229, Wilson C. A., Finger M. H., Harmon B. A., Scott D. M., Wilson R.
doi:10.1051/ 0004-636 1:200809983 B., Bildsten L., Chakrabarty D., Prince T. A. 1997, ApJ, 479, 388,
Ribó M., Negueruela I., Blay P., Torrejón J. M., Reig P. 2006, A&A, 449, doi:10.1086/303841
687, doi:10.1051/0004-6361:20054206 Winn J. N., et al. 2011, AJ, 141, 63, doi:10.1088/0004-6256/141/2/63
Rodriguez J., et al. 2006, MNRAS, 366, 274, doi:10.1111/j.1365- Witte M. G. Savonije G. J. 2002, A&A, 386, 222, doi:10.1051/0004-
2966.2005.09855.x 6361:20020155
Royer F., Grenier S., Baylac M.-O., Gómez A. E., Zorec J. 2002, A&A, 393, Woo J. W., Clark G. W., Levine A. M., Corbet R. H. D., Nagase F. 1996, ApJ,

MNRAS 000, 1–23 (20XX)

A power law relationship in an eccentric system, its role and the physical mechanism of its formation 23
467, 811, doi:10.1086/177655

This paper has been typeset from a TEX/LATEX file prepared by the author.

MNRAS 000, 1–23 (20XX)

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.

daiyongfeng.zip
daiyongfeng.doc