Some properties of doubly-degenerate stars

Shin’ichirou Yoshida∗ and Junya Tanaka

Department of Earth Science and Astronomy, Graduate School of Arts and Sciences, The University of Tokyo,
Komaba 3-8-1, Meguro-ku, Tokyo 153-8902, Japan
(Dated: November 6, 2020)
We investigate critical masses of and circular geodesics around doubly-degenerate stars (DDSs)
which are composed of cold nuclear matter as well as cold Fermionic dark matter (DM). We here
consider asymmetric dark Fermion with self-interaction as a DM candidate. These stars have core-
envelope structures and are categorized into baryon-enveloped and DM-enveloped, according to
the composition of their envelope. It is seen that the baryon-enveloped and DM-enveloped classes
arXiv:2011.02841v1 [astro-ph.HE] 5 Nov 2020

have their own critical masses determined mainly by the dominant component in their envelope.
For a typical parameter sets, we see that a balanced mixture of two species may lead to smaller
masses than if the either of the species is dominant. We also show that for a highly DM-enveloped
case circular orbits in the vacuum region terminates at the innermost stable circular orbit (ISCO) in
vacuum, but circular orbits of smaller radius are possible in the DM-envelope forming a gap between
the ISCO and the inner orbits.

I. INTRODUCTION these exotic objects serve as alternatives to the observed

The nature of cosmic dark matter (DM) still remains
one of the most perplexing mysteries of the physics for
more than 80 years since Zwicky suggested its presence
in clusters of galaxies [1, 2]. The necessity of DM in
galactic scale [3], in clusters of galaxies (e.g., [4]), and in
cosmological scale [5] is now firmly established. Baryonic
astronomical objects that emit too weakly to be detected
are severely constrained as dark matter candidates from
the Big Bang nucleosynthesis. Most promising candi-
dates are yet-to-be-found elementary particles that do
not interact with ordinary matter or at least do very
weakly. [6] From the point of view of the formation of
cosmic structures the dark matter must be subrelativistic
and in the standard theory of DM it is regarded as col-
lisionless (collisionless cold dark matter, CCDM). There
are, however, some astronomical observations that are
at odds with CCDM models. Dwarf galaxies have flat-
ter density profile at their center than is expected from
CCDM [7, 8]. Moreover our Galaxy should have more
subhaloes than are observed as satellite galaxies [9–11].
Also ”Too-big-to-fail” problem [12] exists for the sub-
haloes of our Galaxy to be explained by CCDM. These
suggest that DM may be collisional. One of the pos-
sible modification of CCDM models is to introduce the
self-interaction of DM particles [13–16]. They may nat-
urally solve the issues of CCDM above. One of the cate-
gories that allow the solution is asymmetric DM (ADM,
see [17] for a review). In the early Universe the baryon
to anti-baryon number ratio might be asymmetric and
leads to the baryon dominant Universe as it is now. The
same may hold for DM particles and the rest of the pair-
annihilated may be observed as DM now. If ADM is also
self-interacting it may aggregate to form stellar-sized ob-
jects. This possibility has been studied in exotic models
of compact stars. One of the suggestions made is that
∗ yoshida@ea.c.u-tokyo.ac.jp
black hole candidates. This idea is severely constrained
for accreting objects in X-ray binaries [18].
Objects with typical mass of neutron stars and made
from dark matter are also considered.[19] They are
termed as dark stars [20, 21] and possibility of them to be
alternatives to neutron stars is discussed. On the other
hand, a neutron star may capture DM particles and the
DM accumulated together with nuclear matter (baryon
component) form a new type of compact star. In [22] Le-
ung et al. study the structure of these stars (dark matter
admixed neutron stars, or DANS. See [23] for an alter-
native formation scenario of DANS). This is a baryon-
dominated counter part of dark stars (see also [24] for
their radial stability). We are here interested in the gen-
eral cases that encompass both dark stars and DANS,
which we call as doubly-degenerate stars [25].
In this paper we study characteristics of equilibria com-
posed of baryonic matter and Fermionic dark matter in
complete degeneracy. An equilibrium star is categorized
either to baryon-enveloped star or to DM-enveloped one.
The former has a core composed of baryon and DM which
is covered with an envelope composed of purely baryonic
matter. The latter has instead an envelope composed
of DM. One of our interests is the critical mass beyond
which a stable equilibrium star cannot exists. Although
this has been studied in former studies, we need to pay
a careful attention to find the critical parameter when
a star is composed of multi-component fluid. Another
character of equilibria we study is the circular geodetic
orbits of a test (baryonic) particle. A neutron star in a
X-ray binary have an accretion disk around it whose radi-
ation partly comes from the disk. The disk components
of radiation depends on the accretion state of the disk
itself as well as the inner edge of the disk. For a neutron
star the inner edge is the smaller of either the surface
of the star or the conventional innermost stable circular
orbit (ISCO), which is located at the radius of 6GM/c2
in Schwarzschild coordinate for a mass M of the star.
Unlike neutron stars with radii less than ISCO, we show
 2

for DDS with some parameter sets there appear multiple mass mX are completely degenerated. They have a self-
of stable orbits inside the ISCO. interaction mediated by a boson with the rest mass mφ ,
Possible formation processes of these objects are not whose coupling constant is αX . ǫX and pX are implicitly
clear [20, 23] and beyond the focus of our study here. related by
" #
m4X 2 αX m2X 6
II. FORMULATION ǫX = 3 ξ(x) + 3 x ;
h̄ 9π h̄ m2φ
1 h p 2 (2x2 + 1) − ln(x +
p i
A. Assumptions ξ(x) = x 1 + x 1 + x2)(5),
8π 2
We here study non-rotating stellar equilibria composed and
of baryonic and dark matter whose mutual interaction is " #
negligible. The baryonic matter is a zero-temperature m4X 2 αX m2X 6
pX = 3 χ(x) + 3 x ;
nuclear matter. The dark matter is a self-interacting h̄ 9π h̄ m2φ
Fermion with a vector mediator field. 1
 p 
2 2
 p 
χ(x) = x 1 + x2 x − 1 + ln(x + 1 + x2 )(6),
8π 2 3
B. Tolman-Oppenheimer-Volkoff system for In this paper we focus our interest on the repulsive self-
multi-component stars
interaction of dark Fermions, thus αX > 0. [31] Here x is
the normalized Fermi momentum x = pX /mX . For our
Since the stars in the present study are assumed to numerical computation we solve for v = x2 . The value
be static and spherically symmetric, the spacetime al- of v = v0 at the origin parametrize the Fermi energy of
lows the Schwarzschild coordinate (r, θ, ϕ) in which the DM. It should be remarked that a solution of the so-
spacetime metric is written as, called ’core-cusp’ problem of galactic center may require
ds2 = gµν dxµ dxν = −e2ν dt2 +e2λ dr2 +r2 sin2 θdϕ2 . (1) [29, 32],

Here we assume the geometrized unit, e.g., c = 1 = G.  m  m

φ
−4  α 2
The stress-energy tensor of the matter is written as, 0.1(gcm−3 ) ≤ 1.1× X X
≤ 10(gcm−3 ).
1GeV 10MeV 10−3
(7)
T µν = (ǫ + p)uµ uν + pg µν , ǫ ≡ ǫ B + ǫ X , p ≡ pB + pX .
We use mX = 1GeV, mφ = 10MeV, αX = 10−3 as a
(2)
canonical set of parameters unless otherwise stated.
The energy density ǫ and the pressure p are simple sum-
mations of the baryonic component (with the subscript
B) and of the dark matter (with the subscript X). These 2. Baryonic matter
thermodynamic variables obey independent equations of
state (EOS) which is described below. By using the Ein-
stein’s equation and the extremal conditions of total mass We adopt zero-temperature EOS for neutron
with respect to the matter variables by fixing the baryon star matter by [33, 34] and utilize the fortran
number and the dark Fermion number, we obtain a mod- subroutine nseos1.f provided by the authors at
ified Tolman-Oppenheimer-Volkoff system of equations http://www.ioffe.ru/astro/NSG/NSEOS/. They made
used in [26] (see also [27] and [28] for general cases with analytic fitting formulae for some representative nuclear
a finite interaction between baryonic and dark matter): EOS, e.g., FPS [35], SLy4[36], APR[37]. The stiffest of
The equation for the metric potential ν is the three is APR, which produces the maximum mass
nonrotating neutron star with M ∼ 2.2M⊙ , while the
dν m + 4πpr3 softest is FPS by which a neutron star has a maximum
= , (3)
dr r(r − 2m) mass of M ∼ 1.8M⊙ (see Fig.1).
Rr
where m(r) = 0 ǫ4πr′2 dr′ . The hydrostatic balance of
particle species i = B, X,
III. RESULTS
dpi (ǫi + pi )(m + 4πpr3 )
=− . (4)
dr r(r − 2m) A. Two kinds of equilibria: baryon-enveloped and
dark matter-enveloped

C. Equation of state For the constant v0 sequences, the inclusion of dark

1. Dark matter
As for the EOS of dark matter, we follow the treatment
in [29] (see also [30]). The dark Fermions with the rest
Fermion slightly modify the mass and the radius of the
star from the purely baryonic counterpart when the den-
sity is high enough. These are the branches in the left of
the filled circles on the solid curves in Fig.1. For these
models, the radius of dark matter distribution RDM is
 3

2.25
SLy4,⊙v0 = 0.1 potential of dark Fermion including gravitational contri-
2.00 FPS,⊙v0 = 0.1 bution to energy,
APR,⊙v0 = 0.1
1.75 SLy4,⊙no⊙DM ∂ǫX
FPS,⊙no⊙DM µX = e ν , (8)
∂nX
M(M ⊙ )

1.50 APR,⊙no⊙DM

1.25 which is constant throughout an equilibrium star.

1.00
M 0.082
0.75 1.5 μμ
0.50 1.4 0.080

8 10 12 14 16 18 20 22 1.3 0.078
R(km)

M/M ⊙

μμ
1.2 0.07⊙

FIG. 1. Comparisons of mass-radius relation with/without 0.074

1.1
dark Fermion. Dashed curves are purely baryonic neu-
0.072
tron star sequences for various EOSs. Solid curves are se- 1.0
quences that contain both baryons and dark Fermions. Dark 0.070
Fermion’s momentum parameter is fixed as v0 = 0.1. On 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40
each of the DDS sequence the filled circle marks the bound- v0
ary between baryon-enveloped branch and DM-enveloped one.
The branch with the larger radius corresponds to the DM- FIG. 2. Gravitational mass M of an equilibrium sequence
enveloped. with baryonic mass fixed as MB = 1.02M⊙ . It is plotted as
a function of v0 parameter (solid). Limits of v0 → 0 are nor-
mal neutron star. Also plotted (dashed) are dark Fermion’s
smaller than the radius of the baryon distribution RB , chemical potential µX (in a arbitrary unit) for equilibrium
therefore the stellar radius R = RB . For 0 ≤ r ≤ RDM star as a function of v0 . The circle marks the maximum
the baryonic and dark matter coexist. We call these of M and the triangle marks that of µX . Baryonic EOS
configuration as baryon-dominant. On the righthand of is APR and dark Fermion parameters are (mX , mφ , αX ) =
the circles, the stellar configuration is mainly determined (1GeV, 10MeV, 10−3 ).
by the degenerate pressure of the dark Fermion. The
mass-radius relation largely deviates from that of neu- Baryonic mass of a configuration is defined by the
tron stars. We have RDM > RB and the surface of the proper volume integration of number density of the bary-
star R = RDM . We call it DM-dominant. onic particle nB as,
Z R
M B = mB 4πr2 eλ nB dr, (9)
B. Critical models of radial stability 0

where mB is the mass of baryon particle (nucleon). In

Now we consider the critical mass of the DDS beyond
the same way the dark matter mass is defined by the
which the star becomes unstable. For one parameter se-
number density of dark Fermion nX
quence of equilibrium as cold neutron stars, the critical
mass is the maximum mass as a function of baryon den- Z R
sity (or energy density). With baryonic EOS and the MDM = mX 4πr2 eλ nX dr, (10)
dark Fermion EOS being fixed, the DDS models gener- 0
ally have two parameters that correspond to the energy
where mX is the mass of the dark Fermion. Gravitational
density of baryon and DM at the stellar center. In these
mass M is defined as
multi-parameter system, a (sufficient) stability criterion
for family of equilibria is developed by [38] (see also [39]). Z R
We follow their treatment to find the critical mass of the M= 4πr2 ǫdr, (11)
0
DDS.
We consider a sequence on which the baryonic mass where energy density is the sum of baryonic and dark
mB is kept constant. We may choose λ ≡ v0 as a param- Fermion contribution.
eter to specify a model on the sequence. And we assume In Fig.2 typical behavior of gravitational mass M
an extremum of gravitational mass M exists on the se- and chemical potential µX are shown on a sequence of
quence at λ = λ0 . The theorem I of [38] tells us that MB =constant. At the point dM/dλ = 0 (marked by
if dµ X dM
dλ dλ > 0 on the secment of the sequence around a circle), µX has a negative slope. Therefore models
λ = λ0 , it is unstable in the sense that an equilibrium is with larger λ = v0 satisfy the condition of the instability
dµX dM
not a local minimum of M .[40] Here µX is the chemical dλ dλ > 0. For this baryonic mass we conclude the
 4

2.1 2.1 M
M
DM-fraction DM-fraction
2.0 2.0 0.8
0.8
1.9

MDM/MDM ⊙ MB
MDM/MDM ⊙ MB
1.9
0.6 1.8 0.6

M/M ⊙
M/M ⊙

1.8
1.7
0.4 0.4
1.7 1.6

1.6 0.2 1.5 0.2

1.4
1.5
0.5 1.0 1.5 2.0 2.5 0.25 0.50 0.75 1.00 1.25 1.50 1.75
MB/M ⊙ MB/M ⊙

FIG. 3. (Solid) Maximum gravitational mass of DDS with FIG. 4. Same as Fig.3 except that the baryonic EOS is FPS.
APR as baryonic EOS. (Dashed) Mass fraction of dark matter The dark sector parameters are the same.
MDM
defined as M +M . The dark sector parameters are, mX =
B DM
1.8 M 0.6
1GeV, mφ = 10MeV, αX = 1 × 10−3 . DM-fraction
1.7
0.5
1.6

MDM/MDM ⊙ MB
critical model corresponds to this point. We remark that 1.5 0.4

M/M ⊙
the models with smaller v0 than that of the triangle point 1.4
also satisfies dµX dM
dλ dλ > 0. It should be, however. remem- 1.3 0.3
bered that the stability criterion applies to the part of
the sequence around the extremum of M . Moreover the 1.2
0.2
limit of v0 → 0 corresponds to a pure neutron star model 1.1
with M ∼ 1.2M⊙. This limit is completely stable. We 1.0
conclude this part of the sequence is stable. 0.6 0.8 1.0 1.2 1.4 1.6 1.8
In Fig.3 the critical gravitational mass (solid line) MB/M ⊙
and dark matter fraction (dashed line) defined by
MDM /(MDM + MB ) are plotted as functions of bary- FIG. 5. Same as Fig.3 except the mass of the dark Fermion
onic mass. This is the case with APR EOS for baryons is mX = 2GeV.
and dark Fermion parameters are (mX , mφ , αX ) =
(1GeV, 10MeV, 10−3 ). It is remarkable that the critical
mass is not a monotonic function. When MB is large, we also have small critical mass for DM-enveloped limit.
the DM fraction becomes small and the model is close We expect it because the heavier mediator have smaller
to the pure neutron star. This is the rightmost end of Yukawa radius of interaction and it contribute less to the
the plot. On the other hand for smaller and smaller MB stiffness of the dark Fermion matter.
the star becomes more and more DM-enveloped and the Finally Fig.7 is the same plot as Fig.3 except that αX
critical mass asymptotes to the pure DM star (dark star is larger. This corresponds to the larger repulsive dark-
limit). Between these limit there is a minimum of critical interaction which results in the larger mass of the DM-
mass. For this particular model parameter, nearly equal
amount of contribution from baryon and DM does not 1.8 0.9
M
support a heavy star against its self-gravity. DM-fraction 0.8
The similar characteristics holds for the softer baryonic 1.6
EOS (Fig.4) although the mass supported by the pure 0.7
MDM/MDM ⊙ MB

baryonic matter is smaller. 1.4 0.6

M/M ⊙

When the DM parameters are modified, the charac- 0.5

1.2
teristics may change. In Fig.5 the mass and the DM 0.4
fraction is plotted for APR EOS and (mX , mφ , αX ) = 1.0
2GeV, 10MeV, 10−3 ). This case is compared with Fig.3 0.3
except that the dark Fermion mass is larger. The crit- 0.8 0.2
ical mass is now a monotonic function of MB . This is 0.1
expected from considering the Chandrasekhar limit of a 0.25 0.50 0.75 1.00 1.25 1.50 1.75
star made of Fermion of mass m0 scales as MCh ∝ m−2 0 .
/
MB M ⊙
In Fig.6 we have the plots for the same parameters
except the mediator mass mφ = 50MeV of the dark FIG. 6. Same as Fig.3 except that the mass of the mediator
self-interaction is larger than that of Fig.3. In this case is mφ = 50MeV.
 5

4.0 M be regarded as a conserved total energy of the particle,

DM-fraction
0.8 while the first and the second terms on the right hand
3.5
side are regarded as the kinetic and the potential energy

MDM/MDM ⊙ MB
0.6 Veff . Notice that ǫ includes the rest mass of test particle,
thus it asymptotes to unity as r → ∞.
M/M ⊙

3.0
0.4 0.1
2.5
0.2 0.0
2.0
−0.1
0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50

MB/M ⊙
FIG. 7. Same as Fig.3 except that the coupling αX = 5×10−3 .
enveloped stars.
C. Circular orbits around/in DDS
In this section we study the circular orbits of baryonic
test particles around the DDS. For a baryon-enveloped
model we do not expect to see qualitative difference from
the normal neutron stars, that is, a circular orbit exists
down to the surface of the star if the stellar radius is
larger than 6M , or it is truncated at 6M if the radius is
smaller than 6M . It is, however, not obvious if there ex-
ists stable circular orbit inside the DM envelope of a DM-
enveloped star, when the radius of the star is smaller than
6M . If circular orbits are allowed in the DM-envelope,
accretion disks (or rings) around the DDS may have qual-
itative difference from the standard disk around the nor-
mal neutron stars. Therefore we test the possibility of
circular orbit in DM envelope of DM-enveloped DDS.
Since the spacetime is spherically symmetric, we focus
on the orbit with θ = π/2 without loss of generality.
Let the 4-velocity of the particle to be uµ = dxµ /dτ
that satisfies the normalization uµ uµ = −1 (thus τ is the
proper time). We define two constants of motion, e.g. the
mechanical energy ǫ and the specific angular momentum
ℓ0 ,
ǫ = −gtβ uβ = e2ν ut , ℓ0 = gϕβ uβ = r2 uϕ .
Then the normalization condition of 4-velocity result in
ǫ2 − 1 1

dξ
2 Z For this particular model there are two separated re-
E≡ = + Veff ; ξ ≡
2 2 dτ
and the effective one-dimensional potential Veff is defined
as
e2ν ℓ20
 
1 of the disk seems to extends down to the surface of the
Veff = 2
+ 1 − . (14)
2 r 2
Eq.(13) may be seen as an equation for one-dimensional
motion of a particle in a potential. The left hand side may
Veff
ℓ0 = 1.732
ℓ0 = 2.165
−0.2 ℓ0 = 2.598
ℓ0 = 3.031
−0.3 ℓ0 = √ 12
ℓ0 = 3.897
ℓ0 = 4.33
−0.4
0.0 2.5 5.0 7.5 10.0 12.5 15.0 17.5 20.0
r/M
FIG. 8. Effective potential Veff of reduced radial motion of a
test particle for a star with APR EOS and (mX , mφ , αX ) =
(1GeV, 10MeV, 10−3 ). The star is DM-enveloped with ρc =
2 × 1014 gcm−3 and v0 = 0.4. The gravitational mass M =
2.12M⊙ , baryonic mass mB = 2.40 × 10−4 M⊙ , DM rest mass
MDM = 2.25M⊙ . The radius of the core of baryon and DM
mixture is Rin /M = 0.519 and the radius of the DM envelope
RDM /M = 4.31. The dark-shaded area is the core and the
light-shaded is the envelope whose outer edge is highlighted
by the vertical dashed line. Each curve corresponds to the
different value of specific angular momentum ℓ0 whose values
are expressed in unit of M . Dots mark the local minimum
of the potential at which circular orbit is possible. The star
corresponds to the VISCO.
In Fig.8 we plot Veff for a highly DM-enveloped model
with different value of ℓ0 . Notice that a particle’s radial
motion is represented by E = constant line and E > Veff
corresponds to the allowed region of motion. A bound
orbit must have E < 0. Minima of Veff are radius of cir-
cular orbits. Shaded-regions are the core (dark-colored.
Mixture of baryon and DM) and the envelope (light-
colored. Pure DM). The radius of the star R/M = 4.31
is smaller than VISCO of R/M = 6. In vacuum region
it is well-known that one stable circular orbit exists for
√
ℓ0 > 12M . There is another stable circular orbit in
(12)
the DM-envelope as far as ℓ0 is not so large (ℓ0 ∼ 4M ).
Moreover we √ have a circular orbit in the envelope even
with ℓ0 < 12M .
eν+λ dr, (13) gion where stable circular orbit is allowed. One is in the
vacuum region which terminates at r = 6M . The other
is in the DM-envelop. These regions are detached. Thus
a thin accretion disk around the star may have a gap at
around the surface of the DM-envelope. The inner part
core.
To see where stable circular orbits are allowed we look
at a criterion which results from the curvature of Veff . If
d2 Veff /dr2 at an extremum of Veff is positive, the orbit
 6

is stable. The second derivative is proportional to the 0.10

radial epicyclic frequency squared. In our case the first 0.05
and second derivatives of Veff are, 0.00
ℓ2 ℓ2 −0.05
   
dVeff
= e2ν ν,r 1 + 02 − 03 (15)

Veff
dr r r −0.10 ℓ0 = 0.693
ℓ0 = 1.323
d2 Veff ℓ20 ℓ20 −0.15
   
= 2e 2ν
ν ν 1 + − ℓ0 = 1.952
,r ,r
dr2 r2 r3 −0.20 ℓ0 = 2.582
ℓ0 = 3.212
ℓ2 2ℓ2 ν,r 3ℓ2 ℓ0 = 3.842
   
+e2ν ν,rr 1 + 02 − 03 + 40 , (16) −0.25
ℓ0 = 4.472
r r r −0.30
0 2 4 6 8 10 12 14 16 18
where ν,r and ν,rr represent the first and second deriva- r/M
tives of ν with respect to r. The extrema of the first
derivative are radii of circular orbits which satisfy, FIG. 10. Same as Fig.8 except v0 = 0.1. The gravitational
mass M = 2.12M⊙ , baryonic mass mB = 1.37 × 10−2 M⊙ ,
r3 ν,r DM rest mass MDM = 1.80M⊙ . The radius of the core of
ℓ20 = . (17)
1 − rν,r baryon and DM mixture is Rin /M = 2.68 and the radius of
the DM envelope RDM /M = 9.40.
By nullifying the second derivative and using Eq.(17)
we obtain an equation for critical radius at which a cir-
cular orbit is an inflexion point of V , VISCO. The thick vertical dotted line marks the position
3 of the critical point inside the DM envelope where the
F ≡ ν,rr − 2(ν,r )2 + ν,r = 0 (18) radial stability of circular orbit changes. On the left of
r
the thick dotted line, another region of stable circular
orbit exists. As is seen in the lower panel of the figure,
4 the specific energy the orbit in the range of 2.6 < r/M <
2 6 exceeds that of VISCO. That means a thin circular
accretion disk may be truncated there, since the energy
F

0
of the disk matter must decrease as it accretes inward.
−2
In Fig.10 effective potential is plotted for less DM-
enveloped cases. We have v0 = 0.1 with ρc being fixed.
No VISCO exists in this model. We have only one cir-
1.0 cular orbit for ℓ0 /M ≤ 0.63. A thin accretion disk is ex-
pected to extend down to the core, where thin boundary
ε

0.8 layer may form. Corresponding criterion F and specific

energy ǫ is found in Fig.11. We see that in the star ǫ is a
2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 monotonic function of r and therefore an accretion disk
r/M may extends down to the core without having a gap.

FIG. 9. Criterion F (Eq.(18)) and specific energy ǫ (Eq.(12))

as functions of radial coordinate. The equilibrium model is 60
the same as Fig.8. [Upper panel] Criterion F as a function 40
F

of radial coordinate (solid). The shaded area corresponds to 20

the allowed region of a circular orbit. Thick dotted vertical
0
line marks the inflexion point where F vanishes. The thin
dotted vertical line marks the surface of the star which is
the boundary between DM envelope and the vacuum. [Lower
panel] Specific energy ǫ of the circular orbits. Thick portion of
the curve corresponds to stable circular orbits. The horizontal 0.9
ε

dash-dotted line is the energy at the vacuum last stable orbit

(r = 6M ). 0.8
2 3 4 5 6 7 8 9

The upper panel in Fig.9 shows F of Eq.(18) for a r/M

DM-enveloped configuration in Fig.8. Positive value of F
(shaded area) means a circular orbit is stable. Notice that
the curve crosses zero at r = 6M , which is the vacuum
innermost stable circular orbit (VISCO) around the mass
M . The location of the surface of the DM envelope is
marked by the thin vertical dotted line, which is inside
FIG. 11. Same as Fig.9 Except that v0 = 0.1. The range of r
displayed is within the star.
In Fig.12 we also change ρc as 7 × 1014 gcm−3 . The
DM-envelope is further reduced. It is noted that bary-
 7

onic mass exceeds that of DM for this case. A single baryons. When the envelope is entirely composed of DM,
stable circular orbit exists for each ℓ0 whose radius ex- it is called DM-enveloped. The former may be regarded
tends down to the surface of the core. as DANS and the latter may be regarded as a generalized
dark star. Since an equilibrium state is characterized by
0.050 two parameters, we utilize Sorkin’s general criterion to
0.025 investigate critical mass of an equilibrium sequence be-
0.000 yond which the star becomes unstable. We see that the
baryon-enveloped and DM-enveloped classes have their
−0.025
own critical masses determined mainly by the dominant
Veff

−0.050 ℓ0 = 0.99 component in their envelope. For a typical parameter

ℓ0 = 1.89 sets, we see that a balanced mixture of two species may
−0.075 ℓ0 = 2.789
ℓ0 = 3.689 lead to smaller masses than if the either of the species is
−0.100
ℓ0 = 4.589 dominant. This, however, depends on the mass and the
−0.125 ℓ0 = 5.489
ℓ0 = 6.388
strength of self-interaction of dark Fermion.
−0.150 Another property of equilibrium investigated is the ex-
0 5 10 15 20 25 30 35
r/M istence of stable circular orbits around the star. Espe-
cially interesting is that the envelope may allow geodetic
motion of baryonic gases for DM-enveloped stars. We
FIG. 12. Same as Fig.10 except ρc = 7 × 1014 gcm−3 . The show that for a highly DM-enveloped case circular orbits
gravitational mass M = 0.84M⊙ , baryonic mass MB =
in the vacuum region terminates at VISCO, but another
0.682M⊙ , DM rest mass MDM = 0.214M⊙ . The radius of
the core of baryon and DM mixture is Rin /M = 8.15 and the
region of stable circular orbits are possible in the DM
radius of the DM envelope RDM /M = 9.11. envelope. For more mildly DM-enveloped stars, we have
a single circular orbit for each specific angular momen-
tum down to the baryon-mixed core, at which geodetic
motion is not possible. In this case a thin accretion disk
that forms around the star does not have a gap and ex-
D. Summary tends down to the core. The existence or non-existence
of the gap and the redshift resulting from the gravity in
We explore characteristics of compact stars which is the DM-envelope may affect the spectrum of disk emis-
composed of completely degenerate two species of mat- sion and may be an observable signature of these stars.
ter, i.e., baryons and dark Fermions. Interaction of two Modeling of these emissions may be an interesting astro-
species are neglected. We solve TOV system of equa- physical application of our model.
tions for multi-component matter. The equilibrium mod-
els are parametrized by baryon central density ρc and
Fermi momentum squared v0 of dark Fermion at the ACKNOWLEDGMENTS
center. Equilibrium is classified as baryon-enveloped,
which has a core composed of mixture of baryons and SY was supported by JSPS Grant-in-Aid for Scientific
dark Fermions and has an envelope composed solely of Research(C) 18K03641

[1] F. Zwicky, Die Rotverschiebung von extragalaktischen cosmic structure, Annalen der Physik 524, 507 (2012),
Nebeln, Helvetica Physica Acta 6, 110 (1933). arXiv:1210.0544 [astro-ph.CO].
[2] F. Zwicky, On the Masses of Nebulae and of Clusters of [7] B. Moore, Evidence against dissipation-less
Nebulae, Astrophys. J. 86, 217 (1937). dark matter from observations of galaxy haloes,
[3] V. C. Rubin, J. Ford, W. K., and N. Thonnard, Rota- Nature (London) 370, 629 (1994).
tional properties of 21 SC galaxies with a large range [8] R. A. Flores and J. R. Primack, Observational and
of luminosities and radii, from NGC 4605 (R=4kpc) to Theoretical Constraints on Singular Dark Matter Halos,
UGC 2885 (R=122kpc)., Astrophys. J. 238, 471 (1980). Astrophys. J. Lett 427, L1 (1994).
[4] D. Clowe, M. Bradač, A. H. Gonzalez, M. Marke- [9] G. Kauffmann, S. D. M. White, and B. Guider-
vitch, S. W. Randall, C. Jones, and D. Zarit- doni, The formation and evolution of galax-
sky, A Direct Empirical Proof of the Existence ies within merging dark matter haloes.,
of Dark Matter, Astrophys. J. Lett 648, L109 (2006), Mon. Not. Roy. Astron. Soc. 264, 201 (1993).
arXiv:astro-ph/0608407 [astro-ph]. [10] A. Klypin, A. V. Kravtsov, O. Valenzuela,
[5] V. Springel, C. S. Frenk, and S. D. M. and F. Prada, Where Are the Missing Galac-
White, The large-scale structure of the Uni- tic Satellites?, Astrophys. J. 522, 82 (1999),
verse, Nature (London) 440, 1137 (2006), arXiv:astro-ph/9901240 [astro-ph].
arXiv:astro-ph/0604561 [astro-ph]. [11] B. Moore, S. Ghigna, F. Governato, G. Lake, T. Quinn,
[6] C. S. Frenk and S. D. M. White, Dark matter and J. Stadel, and P. Tozzi, Dark Matter Substructure within

Galactic Halos, Astrophys. J. Lett 524, L19 (1999),
arXiv:astro-ph/9907411 [astro-ph].
[12] M. Boylan-Kolchin, J. S. Bullock, and
M. Kaplinghat, Too big to fail? The puz-
zling darkness of massive Milky Way subhaloes,
Mon. Not. Roy. Astron. Soc. 415, L40 (2011),
arXiv:1103.0007 [astro-ph.CO].
[13] M. Vogelsberger, J. Zavala, and A. Loeb, Sub-
haloes in self-interacting galactic dark matter
haloes, Mon. Not. Roy. Astron. Soc. 423, 3740 (2012),
arXiv:1201.5892 [astro-ph.CO].
[14] M. Rocha, A. H. G. Peter, J. S. Bullock, M. Kapling-
hat, S. Garrison-Kimmel, J. Oñorbe, and L. A. Mous-
takas, Cosmological simulations with self-interacting
dark matter - I. Constant-density cores and sub-
structure, Mon. Not. Roy. Astron. Soc. 430, 81 (2013),
arXiv:1208.3025 [astro-ph.CO].
[15] J. Zavala, M. Vogelsberger, and M. G.
Walker, Constraining self-interacting dark mat-
ter with the Milky way’s dwarf spheroidals.,
Mon. Not. Roy. Astron. Soc. 431, L20 (2013),
arXiv:1211.6426 [astro-ph.CO].
[16] A. H. G. Peter, M. Rocha, J. S. Bullock, and
M. Kaplinghat, Cosmological simulations with self-
interacting dark matter - II. Halo shapes versus obser-
vations, Mon. Not. Roy. Astron. Soc. 430, 105 (2013),
arXiv:1208.3026 [astro-ph.CO].
[17] K. M. Zurek, Asymmetric Dark Matter: Theories,
signatures, and constraints, Phys. Rep. 537, 91 (2014),
arXiv:1308.0338 [hep-ph].
[18] Y.-F. Yuan, R. Narayan, and M. J. Rees, Con-
straining Alternate Models of Black Holes: Type
I X-Ray Bursts on Accreting Fermion-Fermion and
Boson-Fermion Stars, Astrophys. J. 606, 1112 (2004),
arXiv:astro-ph/0401549 [astro-ph].
[19] It should be noted that if DM particles are not asym-
metric as in popular WIMPs (weakly interacting massive
particles) or axions, they may annihilate each other and
heat a neutron star after they are captured by it. These
possibilities have been tested by studying the cooling his-
tory of neutron stars (see, e.g., [41]).
[20] C. Kouvaris and N. G. Nielsen, Asymmetric dark
matter stars, Phys. Rev. D 92, 063526 (2015),
arXiv:1507.00959 [hep-ph].
[21] A. Maselli, P. Pnigouras, N. G. Nielsen, C. Kouvaris, and
K. D. Kokkotas, Dark stars: Gravitational and electro-
magnetic observables, Phys. Rev. D 96, 023005 (2017),
arXiv:1704.07286 [astro-ph.HE].
[22] S.-C. Leung, M.-C. Chu, and L.-M. Lin, Dark-matter ad-
mixed neutron stars, Phys. Rev. D 84, 107301 (2011).
[23] S.-C. Leung, S. Zha, M.-C. Chu, L.-M. Lin, and
K. Nomoto, Accretion-induced Collapse of Dark
Matter Admixed White Dwarfs. I. Formation of
Low-mass Neutron Stars, Astrophys. J. 884, 9 (2019),
arXiv:1908.05102 [astro-ph.HE].
[24] S.-C. Leung, M.-C. Chu, and L.-M. Lin, Equilibrium
structure and radial oscillations of dark matter admixed
neutron stars, Phys. Rev. D 85, 103528 (2012).
[25] These objects are called ’double degenerate stars’ in [42].
We have slightly modified the term because it may be
confused with binary stars whose components are degen-
8
erate stars.
[26] P. Ciarcelluti and F. Sandin, Have neutron stars a
dark matter core?, Physics Letters B 695, 19 (2011),
arXiv:1005.0857 [astro-ph.HE].
[27] M. I. Gresham and K. M. Zurek, Asym-
metric dark stars and neutron star sta-
bility, Phys. Rev. D 99, 083008 (2019),
arXiv:1809.08254 [astro-ph.CO].
[28] T. Kodama and M. Yamada, Theory of Superdense Stars,
Progress of Theoretical Physics 47, 444 (1972).
[29] A. Maselli, P. Pnigouras, N. G. Nielsen, C. Kouvaris, and
K. D. Kokkotas, Dark stars: Gravitational and electro-
magnetic observables, Phys. Rev. D 96, 023005 (2017),
arXiv:1704.07286 [astro-ph.HE].
[30] C. Kouvaris, Limits on Self-Interacting Dark Matter from
Neutron Stars, Phys. Rev. Lett. 108, 191301 (2012),
arXiv:1111.4364 [astro-ph.CO].
[31] Attractive self-interaction does not necessarily mean a
star is unstable to gravitational collapse as shown in [27].
We focus on the repulsive case for simplicity.
[32] S. Tulin and H.-B. Yu, Dark matter self-interactions
and small scale structure, Phys. Rep. 730, 1 (2018),
arXiv:1705.02358 [hep-ph].
[33] P. Haensel and A. Y. Potekhin, Analytical repre-
sentations of unified equations of state of neutron-
star matter, Astron. Astrophys. 428, 191 (2004),
arXiv:astro-ph/0408324 [astro-ph].
[34] A. Y. Potekhin and G. Chabrier, Mag-
netic neutron star cooling and micro-
physics, Astron. Astrophys 609, A74 (2018),
arXiv:1711.07662 [astro-ph.HE].
[35] V. R. Pandharipande and D. G. Ravenhall, Hot Nuclear
Matter, in NATO Advanced Science Institutes (ASI) Se-
ries B, Vol. 205 (1989) p. 103.
[36] F. Douchin and P. Haensel, A unified equa-
tion of state of dense matter and neutron star
structure, Astron. Astrophys. 380, 151 (2001),
arXiv:astro-ph/0111092 [astro-ph].
[37] A. Akmal, V. R. Pandharipande, and D. G. Raven-
hall, Equation of state of nucleon matter and
neutron star structure, Phys. Rev. C 58, 1804 (1998),
arXiv:nucl-th/9804027 [nucl-th].
[38] R. D. Sorkin, A Stability Criterion for Many Parameter
Equilibrium Families, Astrophys. J. 257, 847 (1982).
[39] J. L. Friedman, J. R. Ipser, and R. D. Sorkin, Turn-
ing Point Method for Axisymmetric Stability of Rotating
Relativistic Stars, Astrophys. J. 325, 722 (1988).
[40] In the original argument of [38] a scalar S to be maxi-
mized for an equilibrium state is related to variables E a
as dS = βa dE a . As in [39] we take S = −M and an equi-
librium minimize it. Then we have E a = (−NB , −NX )
and βa = (µB , µX ). Since an equilibrium star satisfies
dM = µB dNB + µX dNX we have dM = µX dNX on the
sequence with baryonic mass being fixed. Thus the sta-
bility condition reads dµ X dM
dλ dλ
> 0.
[41] C. Kouvaris, WIMP annihilation and cooling of neutron
stars, Phys. Rev. D 77, 023006 (2008).
[42] X.-L. Luo, H. Bai, and L. Zhao, GENERAL: Double De-
generate Stars, Chinese Physics Letters 25, 2378 (2008).