Astron. Astrophys. 335, 969-972 (1998)
2. Modified Urca process in strong magnetic fields
Throughout this paper, we consider the magnetic fields beyond
, where G is the critical
magnetic field strength by the relation . Hence,
we must use the relativistic Dirac equation for the electrons, which
yields energy eigenstates
![[EQUATION]](img15.gif)
where quantum number n is given by
![[EQUATION]](img16.gif)
for the Landau level and spin
(For reference, see Mészáros
1993). The usual sum over electron states (per unit volume) in a zero
field is
![[EQUATION]](img19.gif)
which should be replaced as follows when the magnetic field is
nonzero,
![[EQUATION]](img20.gif)
where is the electron Compton wavelength.
Notice that the cyclotron frequency of protons is about 1836
( ) times smaller than the electron cyclotron
frequency. However, as shown by Lai and Shapiro (1991), the values of
for electrons and protons are essentially the
same for n-p-e system because of charge neutrality, here
is the upper limit of the summation over
n in Eq. (6). Therefore, we cannot ignore magnetic field
effects on the proton of state when is
small.
In the following calculations, we ignore the contribution of the
proton branch of the modified Urca process in a magnetic field because
the proton number density is about ten percent of neutron one in
neutron star cores. According to Friman and Maxwell (1979), Yakovlev
and Levenfish (1995) ( ), the neutrino energy
production rate of the neutron branch in zero
magnetic field is given by
![[EQUATION]](img27.gif)
where (j=1-4) are the nucleon momenta and
is the electron momentum,
are the corresponding Fermi momenta of
particles, is the squared matrix element of the
first process (2), s=2 is a symmetry factor for that process, T
is the temperature of the equilibrium system,
is the dimensionless momenta of neutrino and
are those of other particles, here is the Fermi
velocity and is an effective mass and
is the Fermi-Dirac function. The integrations
in A and I are standard and yield (e.g. Shapiro &
Teukolsky 1983)
![[EQUATION]](img37.gif)
Taking the nucleon-nucleon interactions into account, Friman and
Maxwell (1979) obtained
![[EQUATION]](img38.gif)
The previous studies demonstrated that the matrix elements for the
neutron decay and electron capture reactions when magnetic field is
nonzero are the same as those when magnetic field is zero
(Fassio-Canuto 1969, Canuto & Ventura 1977, Lai & Shapiro
1991). We therefore need only consider the effect of a magnetic field
on phase-space integration of the charged particles. Hence the energy
loss rate of the modified Urca process in a magnetic field is written
as
![[EQUATION]](img39.gif)
where is the rate in zero magnetic field,
while R is the factor which describes the effect of the
magnetic field on the modified Urca process. This factor can be
written as follows according to Eqs. (8) and (9):
![[EQUATION]](img41.gif)
After integrating over with j=1-3, we
have
![[EQUATION]](img43.gif)
where
![[EQUATION]](img44.gif)
and in Eq. (16)
are the magnetic field strengths in units of the critical magnetic
field for electrons and protons, respectively.
Fig. 1 shows the ratio R as a function of density
when T= K and
T= K, respectively. The saw-teeth shape of the
curves is due to the fact for B , we have more
available phase space for electrons at given density, which causes the
decay rate to increase. For the same reason, a nonzero magnetic field
decreases the inverse -decay rate (Lai &
Shapiro 1991), the competition of these two reaction rates therefore
causes the energy loss rate to oscillate with the increase of density.
The stronger the magnetic field is, the larger the amplitude is. But
the tendency is to increase the reaction rate of the modified Urca
process in the range of the core density. It is also evident from
Fig. 1 that the temperature only influences the reaction rate
when the electrons are in resonant states at which the electrons begin
to fill the higher Landau levels.
![[FIGURE]](img50.gif) |
Fig. 1. Ratio R versus the density of neutron star cores under the influence of different interior magnetic field strengths
|
© European Southern Observatory (ESO) 1998
Online publication: June 26, 1998
helpdesk@link.springer.de  |