Normalized to: Soni, V.
[1]
oai:arXiv.org:1804.05343 [pdf] - 1889312
Magnetar Crusts: Matter at $10^{15}$ Gauss
Submitted: 2018-04-15, last modified: 2019-05-26
In this work we find that the at the high polar magnetic fields of magnetars,
$B \sim 10^{14-15}$ G, the outermost crust ($\rho < 10^7$ gm/cc) of the star
can become a transverse insulator and a filamentary crystal along the field
direction. Also, the transverse conductivity in the crust goes inversely as the
square of polar magnetic field (as $1/B^2$). At these high fields the
transverse crustal currents associated with the polar magnetic field can then
dissipate more effectively via Ohm's law.
[2]
oai:arXiv.org:1805.09770 [pdf] - 1779541
Magnetar Signature - The U Curve
Submitted: 2018-05-24
This work looks at some definitive signatures of magnetars, in particular of
period closures accompanied by a decline of X-ray radiation in two models. We
review some of the previous works which are based on the well known dynamo
model in which the star is born with a period of a few milliseconds at high
temperatures. In such a convection regime the dynamo mechanism can amplify the
the magnetic fields to the magnetar value. This is in contrast to a screened
core model which posits that a high density phase transition occurs in the
inner core of magnetars that dynamically aligns all the neutron magnetic
moments producing a large magnetic field in the core. The accompanying change
of flux gives rise to shielding or screening currents in the surrounding high
conductivity plasma that do not permit the field to exit to the surface.
Ambipolar diffusion then transports the field to the crust dissipating energy
in neutrinos and X-rays. The up-welling field cleaves the crust resulting in
flares and X-ray radiation from ohmic dissipation in the crust till the
screening currents are spent and the surface polar field attains its final
value. In the dynamo model the polar magnetic field decreases with time whereas
in our screened model it increases to its final value. One consequence of this
is that in the latter model, as a function of time and period, the ratio of the
dipole radiation loss, $\dot E$ to the X-ray luminosity, $ L_X$, is a 'U'
curve, indicating that it is the exponential decline in $L_X$, that brings
closure to the periods that are observed for magnetars.
[3]
oai:arXiv.org:1509.00706 [pdf] - 1270738
The cosmological constant from the zero point energy of compact
dimensions
Submitted: 2015-09-02
We consider extra compact dimensions as the origin of a cosmological
universal energy density in the regular dimensions, with only graviton fields
propagating in the compact space dimensions. The quantum zero point energy
originating from the finite size boundary condition in the compact dimensions
can produce a constant energy density in regular $3$ space which is homogeneous
and isotropic. It then makes a natural identification with the cosmological
constant in conformity with the Einstein equation. It turns out that for the
emergent energy density to agree with the observed value of the cosmological
constant, the size/radius of compact dimension is to be of order of $10^{-2}$
cm.
[4]
oai:arXiv.org:1407.0842 [pdf] - 846506
The nucleon and the two solar mass neutron star
Submitted: 2014-07-03
The existence of a star with such a large mass means that the equation of
state is stiff enough to provide a high enough pressure up to a fairly large
central densities,. Such a stiff equation of state is possible if the ground
state has nucleons as its constituents. This further implies that a purely
nucleon ground state may exist till about four times nuclear density which
indicates that quarks in the nucleon are strongly bound and that the nucleon
nucleon potential is strongly repulsive. We find this to be so in a chiral
soliton model for the nucleon which has bound state quarks. We point out that
this has important implications for the strong interaction $ \mu_B$ vs T phase
diagram.
[5]
oai:arXiv.org:1211.1185 [pdf] - 586173
Magnetic Moment of Constituent Fermions in Strongly Interacting Matter
Submitted: 2012-11-06
We investigate the magnetic moment operator for constituent fermion masses
for chirally symmetric theories. Constituent fermion masses are generated
through a yukawa interaction of the fermion with a scalar (or /and
psuedoscalar) field via the vacuum expectation value (VEV) of the scalar (or
and psuedoscalar) field. We especially consider the high baryon density $\pi_0$
condensed phase, in which chiral symmetry is spontaneously broken, with space
varying expectation values of the $ \sigma$ and $\pi_0$ fields. This phase has
a spin polarized fermi sea as the ground state. We show that there is indeed
generated a macroscopic magnetization in this phase, contrary to what one would
have found, if one just used a primitive phenomenological magnetic moment
formula for explicit/ current fermion masses. Furthermore, this analysis
reveals that the magnetization of this state goes up as the VEV, that
determines the 'mass', comes down with increasing baryon density. The
consequent high magnetic field that is generated will destabilize this state at
a threshold density. This is important in the context of neutron stars, as such
a high density state may be responsible for very high magnetic fields in the
dense core of neutron stars. This could potentially be the origin of magnetars
- the stars with the largest magnetic fields in the universe.
[6]
oai:arXiv.org:1111.7092 [pdf] - 586028
Is the high mass binary pulsar PSR J 1614-2230 a latent magnetar?
Submitted: 2011-11-30, last modified: 2012-11-06
We consider the newly found high mass and low magnetic field binary pulsar
PSR J1614-2230 in a model in which magnetars owe their strong magnetic fields
to a high baryon density, magnetized core. In our magnetar model all neutron
stars above a certain threhold mass are magnetars. This confronts us with the
very basic paradox as to why this high mass star, PSR J1614-2230, remains a
pulsar and shows no magnetar characteristics. This is a star that has acquired
its large mass by accretion from its binary companion over 5 gigayears.
In this work we show that the maximum rate of energy gain from the strong
interaction phase transition from this slow accretion does not allow for high
enough interior temperature for ambipolar transport of the magnetic field to
the surface of the star and thus the PSR J 1614-2230 remains latent and does
not become an emergent magnetar.
[7]
oai:arXiv.org:1012.1420 [pdf] - 1042413
Magnetars From Magnetized Cores Created by a Strong Interaction Phase
Transition
Submitted: 2010-12-07
We consider a model where the strong magnetic fields of magnetars arise from
a high baryon density, magnetized core. In this framework magnetars are
distinguished from pulsars by their higher masses and central density. For
magnetars, as core densities exceed a threshold, the strong interaction induces
a phase transition to a neutral pion condensate that aligns all magnetic
moments. The core magnetic field is initially shielded by the ambient high
conductivity plasma. With time the shielding currents dissipate transporting
the core field out, first to the crust and then breaking through the crust to
the surface of the star. Recent observations provide strong support for this
model which accounts for several properties of magnetars and also enables us to
identify new magnetars.
[8]
oai:arXiv.org:0705.0592 [pdf] - 998
A Natural Explanation for Magnetars
Submitted: 2007-05-04
We explore the possibility that a magnetar may owe its strong magnetic field
to a magnetized core which, as indicated by certain equations of state, may
form due to phase transitions at high density mediated by strong interaction
within a sufficiently massive neutron star. We argue that the field derived
from such a core could explain several inferred evolutionary behaviors of
magnetars.