Документ взят из кэша поисковой машины. Адрес оригинального документа : http://xray.sai.msu.ru/~polar/html/publications/sakharov2002/s3_popov.ps Дата изменения: Wed Nov 6 19:48:19 2002 Дата индексирования: Sat Dec 22 05:14:29 2007 Кодировка:

YOUNG COMPACT OBJECTS IN THE
SOLAR VICINITY
S.B. Popov, 1;4 M.E. Prokhorov, 1 M. Colpi, 2 A. Treves, 3
and R. Turolla 4
1 Sternberg Astronomical Institute, Universitetski pr. 13, Moscow 119992, Russia,
2 University of Milano-Bicocca, Piazza della Scienza 3, Milano 20126, Italy,
3 University dell'Insubria, via Vallegio 11, Como 22100, Italy,
4 University of Padova, via Marzolo 8, Padova 35131, Italy
We present Log N { Log S distribution for close-by young isolated neutron stars.
On the basis of this distribution it is shown that the seven ROSAT isolated
neutron stars (if they are young cooling objects) are genetically related to the
Gould Belt. We predict, that there are about few tens unidentied close-by young
isolated neutron stars in the ROSAT All-Sky Survey. The possibility that these
seven peculiar sources contain a neutron star less massive and more magnetized
then in ordinary radiopulsars is also discussed. In the aftermath of relatively
close recent supernova explosions (1 kpc around the Sun, a few Myrs ago), a few
black holes might have been formed, according to the local initial mass function.
We thus discuss the possibility of determining approximate positions of close-by
isolated black holes using data on runaway stars and simple calculations of binary
evolution and disruption.
1 Introduction
Neutron stars (NSs) and black holes (BHs) are among the most interesting
astrophysical sources. Usually NSs are observed as radio pulsars or as accreting
objects in close binaries. Similarly, stellar mass BHs are observed when they
accrete matter from a companion star. Here we focus on much more elusive
sources, namely isolated NSs (which may show no radiopulsar activity) and
isolated BHs.
An isolated NS can be relatively bright in soft X-rays due to its thermal
emission during the rst Myrs of its life, when it is still hot (T  10 6 K) in
the aftermath of the supernova (SN) explosion. Such objects are observed in
the Solar proximity and in SN remnants [1]. Older NSs (that is to say those
which crossed the deathline in  10 7 yr) are not expected to emit appreciable
amounts of electromagnetic radiation in any energy band. However, accretion
of the interstellar medium (ISM) may make them shine again as soft, faint
X-ray sources (see e.g. Treves et al. [2]). Much in the same way, an isolated

BH may be detected if it accretes from the ISM, or, possibly, revealed through
microlensing [3].
In this paper we discuss the possible origin of close-by, isolated NSs and
present some evidence that the seven radio-quiet, thermally emitting ROSAT
sources (the \magnicent seven" [2]) may be characterized by dierent values
of the stellar parameters (mass and magnetic eld) with respect to ordinary
radiopulsars. We rst construct the Log N { Log S distribution for young close-
by isolated NSs and compare it with present observations of close-by young
isolated NSs of all types. Then we discuss how the alignment of the magnetic
and rotation axes, together with the role of fall-back following the supernova
event, may help in explaining the observed parameters of NSs. Finally in x3
we discuss how one can estimate an approximate positions of close-by young
isolated BHs in the light of their possible detection at X/gamma-rays energies.
2 Isolated neutron stars
In this section we discuss isolated NSs. The material presented here is partly
based on the results published in Popov et al. [4]. In addition some recent
results are also included [5].
2.1 Origin of close-by isolated NSs
To understand how the local population of isolated NSs originated, we con-
struct their Log N { Log S distribution. The main components of our model
are (see [5]): spatial distribution of NS progenitors, NS formation rate, NS
cooling history, and a model of interstellar absorption (that is to say the spa-
tial distribution of the ISM). In addition we calculate the dynamical evolution
of NSs in the galactic potential. In brief our model can be described in the
following way: NSs are born in the Galactic plane and in the Gould Belt (a
local compound of stellar associations, see below); at birth they receive a kick
velocity; we then follow the evolution of NSs in the Galactic potential; nally,
we calculate the ROSAT count rate basing on cooling curves and an assumed
model of interstellar absorption.
NSs are considered to be born with a constant rate: 20 NSs per Myr come
from the Gould Belt, and 250 NSs per Myr from the Galactic plane (up to
a limiting distance of 3 kpc from the Sun) with a uniform distribution. The
Gould Belt is modeled as a disk of 500 pc radius with an inclination of 18 ф
with respect to the Galactic plane. Its center is situated at 100 pc from the

Sun in the Galactic anticenter direction. The central region (150 pc in radius)
is devoided of newborn NSs (see Poppel [6] and Torra et al. [7]).
Log S, cts/s
­2 ­1 0 1
­1
0
1
2
RBS
RX 1856
Vela
0659+14
RX 0720
RBS 1556
Geminga
1057­52
RX 0806
RBS 1223
RBS 1774
RX 0420
3EG 1835
1920+10
Figure 1: All-sky Log N - Log S distribution. Black triangles { the seven
RINSs; crosses { Geminga, \three musketeers", 1929+10 and 3EG J1835. We
also show the ROSAT Bright Sources (RBS) limit (Schwope et al. [10]). Upper
curve: NSs born in the Gould Belt and in the Galactic disk (r disk = 3 kpc,
total birth rate 270 Myr 1 ). Lower curve: NSs born only in the Galactic disk
(r disk = 3 kpc, birth rate 250 Myr 1 ).
To calculate the thermal evolution of NSs we use the data obtained by
Sankt-Petersburg group (see Kaminker et al. [8], and the review by Yakovlev
et al. [9]). The NS cooling depends on the star mass and we adopt a at mass
spectrum in the range 1:1 M < M < 1:8 M . A more standard spectrum with
a sharp maximum around 1.35-1.4 M gives nearly the same result. Cooling
curves take into account all neutrino processes but ignore neutron super uidity
in the crust and core since this is not expected to in uence the nal results

signicantly. Calculations for each NS are truncated when its temperature
drops to 10 5 K; this corresponds to a NS age of 4.25 Myrs for the lightest NSs
(M = 1:1 M ) or less for more massive objects.
Since we expect the NS to emit most of its luminosity at UV/soft X-ray
energies ( 20 200 eV or T  10 5 {10 6 K) interstellar absorption plays a cru-
cial role as far as the observability of these sources is concerned. Any attempt
to estimate the amount of observable cooling isolated NSs using unabsorbed
ux greatly overestimates their number.
Our main results are presented in Fig. 1 where we compare the Log N
{ Log S for NSs born in the Gould Belt and the Galactic disk. All curves
refer to the whole sky. As can be seen the contribution from NSs born in the
disk fails to explain the observed distribution while the inclusion of the objects
originating from the Gould Belt alone can match the observations. Absorption,
the at geometry of NS initial distribution and the nite extension of the Belt
naturally explain the very at (slope < 1) Log N { Log S distribution.
Our calculations show that there may be at most a few dozens of uniden-
tied close-by isolated NSs in the ROSAT All-Sky Survey (at count rate >
0.015 cts s 1 ) depending on parameters of the model. Also there may be a
few unidentied ROSAT isolated NSs (RINSs) with uxes > 0:1 cts s 1 at
low Galactic latitudes (see also Schwope et al. [10]). Most objects should
be observed at jbj < 20 ф towards the directions of lower absorption. Some
of them can have counterparts among unidentied gamma-ray sources (also
possibly connected with the Gould Belt, see Grenier [11]). Identication of
these objects can be important for choosing a correct cooling model and for
determination of the mass spectrum of NSs.
2.2 Census of close-by young NSs
At present about 20 NSs are known which are younger than 4.25 Myrs and
closer than 1 kpc to the Sun (see the Table). They include: the \magni-
cent seven" (radio-quiet, ROSAT isolated NSs with only thermal emission),
Geminga and the Geminga-like object 3EG J1835 (pulsars the beams of which
do not intersect the Earth), the \three musketeers" (Vela, PSR 0656+14, PSR
1055-52), PSR 1929+10 and seven young radio pulsars, which have not been
detected in X-rays yet.
In addition to the observed sources, we expect about one hundred isolated
NSs younger than 4 Myr inside 1 kpc. These NSs are not detected as radio
pulsars, but tens of them can be identied in ROSAT data as dim sources

(others are too old to be hot enough). Pulsar beaming can be responsible only
for a fraction of these undetected (in the radio) young NSs (about 50-70% of
young pulsars are not visible from Earth [12]), and most of RINSs should be
really radio silent. This provides strong support to the arguments by Gotthelf
and Vasisht [13], that \at least half of the observed young neutron stars follow
an evolutionary path quite distinct from that of the Crab pulsar".
2.3 Are RINSs of a dierent stock ?
An interesting feature of RINSs population is the detection of periods in the
 10-20 s range (typical of SGRs/AXPs, i.e. of magnetars) for four objects.
Present data allow to exclude any pulsation (down to a few % fraction) for at
least one source, RX J1856.5-3754. V. Beskin (2001, private communication)
suggested this could be due to the alignment of magnetic and spin axes (see
for example Tauris & Manchester [12] for a discussion). Alignment is a process
which leads to \period freezing" and a low pulsed fraction.
However in the case of coolers alignment should operate on short timescale,
since the star cools down in  1 Myr. For radio pulsars the timescale of align-
ment is about 10 Myrs or longer [12], so it seems unlikely that this mechanism
is responsible for RINSs distribution of the pulsed fraction, unless one assumes
that RINSs form a separate population from normal radio pulsars. To illustrate
this let us assume that the alignment timescale is  align
/(
2
0
cos 2  0 B 2
0
) 1
(here  0
and B 0
are initial values of an angle between spin and magnetic axis
and of magnetic eld). The previous expression comes from magnetodipolar
braking supplemented by the
condition
0
cos 0
=
cos. To explain the
dierence between  align in RINSs and radio pulsars RINSs need to have a
dierent distribution in B 0 and/or  0 . In this respect RINSs may come from
the same population as radiopulsars but are characterized by dierent average
properties, like e.g. higher values of the magnetic eld and relatively higher
surface temperatures. The latter would imply a lower mass for NSs of the
same age.
RINSs are currently thought to be rather highly magnetized objects (in RX
J0720.4-3125 the detected spind-down implies B  2  10 13 G [15]). If this
is indeed the case, then one has to explain why their B-eld is a factor  10
higher than the average value in radiopulsars ( 2  10 12 G). A possibility is
that NSs with higher magnetic elds are hotter. It is known that less massive
NSs cools more slowly because direct URCA processes are not eective (see
e.g. [9]). This means that among NSs of the same age the lighter are the
hotter, so to test our hypotesis we need to show that lighter NSs may support

Table 1: Local (r < 1 kpc) population of young (age < 4:25 Myrs) isolated
neutron stars.
Object name Period, CR a , _
P Dist., Age b , Ref.
s cts/s =10 15 kpc Myrs
RX J1856.5-3754 | 3.64 | 0.117 e  0:5 [2, 14]
RX J0720.4-3125 8.37 1.69  30 60 | | [2, 15]
RX J1308.6+2127 10.3 0.29 < 10 4 ? | | [2, 16, 17]
RX J1605.3+3249 | 0.88 | | | [2]
RX J0806.4-4123 11.37 0.38 | | | [2, 18]
RX J0420.0-5022 22.7 0.11 | | | [2]
RX J2143.7+0654 | 0.18 | | | [19]
PSR B0633+17 0.237 0.54 d 10.97 0.16 e 0.34 [20]
3EG J1835+5918 | 0.015 | | | [21]
PSR B0833-45 0.089 3.4 d 124.88 0.294 e 0.01 [20, 22, 23]
PSR B0656+14 0.385 1.92 d 55.01 0.762 f 0.11 [20, 23]
PSR B1055-52 0.197 0.35 d 5.83  1 c 0.54 [20, 23]
PSR B1929+10 0.227 0.012 d 1.16 0.33 e 3.1 [20, 23]
PSR J0056+4756 0.472 | 3.57 0.998 f 2.1 [23]
PSR J0454+5543 0.341 | 2.37 0.793 f 2.3 [23]
PSR J1918+1541 0.371 | 2.54 0.684 f 2.3 [23]
PSR J2048-1616 1.962 | 10.96 0.639 f 2.8 [23]
PSR J1848-1952 4.308 | 23.31 0.956 f 2.9 [23]
PSR J0837+0610 1.274 | 6.8 0.722 f 3.0 [23]
PSR J1908+0734 0.212 | 0.82 0.584 f 4.1 [23]
a ROSAT count rate
b ) Ages for pulsars are estimated as P=(2 _
P ),
for RX J1856 the estimate of its age comes from kinematical considerations.
c ) Distance to PSR B1055-52 is uncertain ( 0.9-1.5 kpc)
d ) Total count rate (black body + non-thermal)
e ) Distances determined through parallactic measurements
f ) Distances determined with dispersion measure

a stronger eld. Such a correlation arises quite naturally if more massive
NSs get their additional mass from fall-back. In this case their magnetic eld
can be signicantly suppressed [24], so more massive NSs should have lower
initial magnetic elds. Besides, strong initial magnetic eld together with fast
rotation can prevent strong fall-back (this is especially possible if the magneto-
rotational mechanism of supernova explosion is valid, see [25, 26]). Again this
leads to the same correlation between mass and led strength, i.e. NSs with
stronger elds would have lower masses. Also, this picture makes room for
long initial spin periods. The study of this (and possible other) correlation in
isolated NSs may prove very useful in understanding the correct mechanism
of the SN event which gave birth to these objects.
In summarizing x2, we stress that future determination of RINSs parallax
and proper motion may help in tracing back their kinematical history and
derive their age. It can give a clue to their mass determination basing on
cooling curves (see Kaminker et al. [8]). Our results suggest that the fraction
of low-mass NSs (M  < 1.3 M ) may not be small. On the other hand there
should be room for NSs with M  > 1:4M , because otherwise the number of
bright objects would be too large.
3 Close-by young isolated black holes
In this section we base on the results published in Popov et al. [4] and
Prokhorov and Popov [27].
SNae explosions produce not only NSs, but also BHs. It is commonly
accepted that BHs are one order of magnitude less abundant than NSs. This
estimate comes from the critical mass for BH formation and follows if one
assumes that progenitors more massive than about 35 M ended as BHs.
Having dozens of SNae in the close solar vicinity during the last 10 Myr we
can expect several BHs to have formed during the same period in the solar
neighborhood.
At present, 56 runaway stars are known within  700 pc from the Sun[28].
Only a few of them result from star-star interactions, so the majority comes
from SNae explosions in binary systems. If the above considerations are correct
we can expect about 5 BHs formed in about 50 disrupted binaries.
Close-by massive runaway stars give us a chance to calculate the approxi-
mate position of close-by young isolated BHs. Among runaway stars the most
massive are  Cep,  Pup, HIP 38518 and  Per (see Hoogerwerf et al. [28]).
Since their mass is  > 33 M , the companion (actually the primary in the orig-

inal binary) was even more massive on the main sequence stage. So, the most
likely product of the explosion of such a massive star should be a BH.
If the present velocities of runaway stars are known, one can estimate their
ages and places of birth. This has been done by Hoogerwerf et al. [28]. To
calculate the present position of a BH we have to know the binary parameters,
i.e., the masses of stars before the explosion, the BH mass, the eccentricity
of the orbit before the explosion, the orbit orientation, and nally the kick
velocity of the BH. Some parameters can be inferred from the observation
of the secondary star. We can assume a zero kick velocity for BHs and zero
orbital eccentricity. Other parameters should be varied within assumed ranges
(see details in Prokhorov and Popov [27].
We calculated approximate positions of isolated BHs for the four systems
mentioned above and estimated error boxes where these BHs could be found.
For  Per and  Pup we obtained not very large error boxes inside each of
which only one unidentied EGRET source is known. We suggest that these
objects can be young isolated BHs. For the two other systems (HIP 38518
and  Cep) the present position of the BH is more uncertain and no denite
conclusion on the possible detectability of the collapsed object can be drawn.
4 Conclusions
We have presented evidence that the seven radio-quiet ROSAT isolated NSs
discovered so far can be connected with recent SNae explosions in the Gould
Belt. These events produced nearby runaway stars and peculiar features in the
local ISM including the Local Bubble. The relatively high local spatial density
of young NSs is a natural consequence of the large number of massive progen-
itors in the Belt. The lack of a similar overabundance of active radiopulsars
in the Solar vicinity lends further support to the claim that a large fraction
( 50% or more) of young NSs should be radio quiet. According to our re-
sults, the ROSAT All Sky Survey may contain about a few tens of unidentied
RINSs. Moreover, it is possible that some unidentied RINSs with quite large
ux (> 0:1 cts s 1 ) are still hiding at low Galactic latitudes.
We also propose that massive runaway stars may be used to trace the
present position of young close-by isolated BHs. Our calculations allowed to
estimate with reasonable accuracy the positions of four such BHs. In two cases
the error box is not too large and this may lead in the future to the positive
identication of an isolated BH with X/gamma-rays observations.

5 Acknowledgments
We want to thank D.G. Yakovlev for the data on cooling curves and comments
on them, and V.S. Beskin, M. Chieregato, A. Possenti and L. Zampieri for dis-
cussions. The work of S.P. was supported by the RFBR grant 02-02-06663 and
by RSCI; that of M.P.by RFBR grant 01-15-99310. S.P. thanks Universities
dell'Insubria and Milano-Bicocca for hospitality.
[1] W. Becker and G. Pavlov, in: \The Century of Space Science", eds. J.
Bleeker, J. Geiss and M. Huber, Kluwer Academic Publishers (in press)
(astro-ph/0208356) (2002).
[2] A. Treves, R. Turolla, S. Zane and M. Colpi, PASP 112, 297 (2000).
[3] E. Agol, M. Kamionkowski, L.V.E. Koopmans and R.D. Blandford, ApJ
576, L131 (2002).
[4] S.B. Popov, M.E. Prokhorov, M. Colpi, A. Treves and R. Turolla, Gravi-
tation & Cosmology (in press) (astro-ph/0201030) (2002).
[5] S.B. Popov, M. Colpi, M.E. Prokhorov, A. Treves and R. Turolla, in
preparation (2003).
[6] W. Poppel, Fund. Cosm. Phys. 18, 1 (1997).
[7] J. Torra, D. Fernandez and F. Figueras, A&A 359, 82 (2000).
[8] A.D. Kaminker, D.G. Yakovlev and Y.O. Gnedin, A&A 383, 1076 (2002).
[9] D.G. Yakovlev, K.P. Levensh and Yu.A. Shibanov, Physics-Uspekhi 42,
737 (1999).
[10] A.D. Schwope, G. Hasinger, R. Schwarz, F. Haberl and M. Schmidt, A&A
341, L51 (1999).
[11] I.A. Grenier, A&A 364, L93 (2000).
[12] T.M. Tauris and R.N. Manchester, MNRAS 298, 625 (1998).
[13] E.V. Gotthelf and G. Vasisht, in: Proceedings of IAU Coll. 177 \Pulsar
astronomy - 2000 and beyond", eds. M. Kramer, N. Wex, and N. Wielib-
inski, ASP Conf. Series 202, 699 (2000).

[14] D.L. Kaplan, M.H. van Kerkwijk and J. Anderson, ApJ 571, 447 (2002).
[15] S. Zane, F. Haberl, M. Cropper, V. Zavlin, D. Lumb, S. Sembay and C.
Motch, MNRAS 334, 345 (2002).
[16] V. Hambaryan, G. Hasinger, A.D. Schwope and N.S. Schulz, A&A 381,
98 (2001).
[17] F. Haberl, Talk presented at the Cospar Meeting held in Houston (2002).
[18] F. Haberl and V.E. Zavlin, A&A 391, 571 (2002).
[19] L. Zampieri et al., A&A 378, L5 (2001).
[20] W. Becker and J. Trumper, A&A 326, 682 (1997).
[21] N. Mirabal and J.P. Halpern, ApJ 547, L137 (2001).
[22] G.G. Pavlov, V.E. Zavlin, D. Sanwal, V. Burwitz, and G.P. Garmire, ApJ
552, L129 (2001).
[23] ATNF Pulsar Catalogue,
http://wwwatnf.atnf.csiro.au/research/pulsar/catalogue/
[24] D. Page, U. Geppert and T. Zannias, Mem. Soc. Astr. It. 69, 1037 (1998).
[25] G.S. Bisnovatyi-Kogan, AZh 14, 652 (1970).
[26] M.E. Prokhorov and K.A. Postnov, Odessa Astr. Publ. 14, 78 (astro-
ph/0110176) (2001).
[27] M.E. Prokhorov and S.B. Popov, Astronomy Letters 28, 542 (2002).
[28] R. Hoogerwerf, J.H.J. de Bruijne and P.T. de Zeeuw, A&A 365, 49 (2001).