Astron. Astrophys. 337, 96-104 (1998)
2. Tidal torques
The explanation of galaxies spins gain through tidal torques was
pioneered by Hoyle (1949) in the context of a collapsing protogalaxy.
Peebles (1969) considered the process in the context of an expanding
world model showing that the angular momentum gained by the matter in
a random comoving Eulerian sphere grows at the second order in
proportion to (in a Einstein-de Sitter
universe), since the proto-galaxy was still a small perturbation,
while in the non-linear stage the growth rate of an oblate homogeneous
spheroid decreases with time as .
More recent analytic computations (White 1984; Hoffman 1986 and
R88a) and numerical simulations (Barnes & Efstathiou 1987) have
re-investigated the role of tidal torques in originating galaxies
angular momentum. In particular White (1984) considered an analysis by
Doroshkevich (1970) that showing as the angular momentum of galaxies
grows to first order in proportion to t and that Peebles's
result is a consequence of the spherical symmetry imposed to the
model. White showed that the angular momentum of a Lagrangian sphere
does not grow either in the first or in the second order while the
angular momentum of a non-spherical volume grows to the first order in
agreement with Doroshkevich's result.
Hoffman (1986) has been much more involved in the analysis of the
correlation of the growth of angular momentum with the density
perturbation . He found an angular
momentum-density anticorrelation: high density peaks acquire less
angular momentum than low density peaks. One way to study the
variation of angular momentum with radius in a galaxy is that followed
by R88a. In this approach the protogalaxy is divided into a series of
mass shells and the torque on each mass shell is computed separately.
The density profile of each proto-structure is approximated by the
superposition of a spherical profile, , and a
random CDM distribution, , which provides the
quadrupole moment of the protogalaxy. To the first order, the initial
density can be represented by:
![[EQUATION]](img30.gif)
where is the background density and
is given by:
![[EQUATION]](img33.gif)
being the power spectrum, while the density
profile is (Ryden & Gunn 1987):
![[EQUATION]](img35.gif)
![[EQUATION]](img36.gif)
where is the height of a density peak,
is the two-point correlation function,
and are two spectral
parameters (BBKS, Eq. 4.6a, 4.6d) while is a
function given in BBKS (Eq. 6.14). As shown by R88a the net rms torque
on a mass shell centered on the origin of internal radius r and
thickness is given by:
![[EQUATION]](img42.gif)
![[EQUATION]](img43.gif)
where , the multipole moments of the shell
and , the tidal moments, are given by:
![[EQUATION]](img46.gif)
![[EQUATION]](img47.gif)
![[EQUATION]](img48.gif)
where is the mass of the shell,
and are the spherical
Bessel function of first and second order while the power spectrum
is given by:
![[EQUATION]](img52.gif)
![[EQUATION]](img53.gif)
(Ryden & Gunn 1987). The normalization constant A can be
obtained, as usual, by fixing that the mass variance at
, that is , be equal to
unity. Filtering the spectrum on cluster scales,
, we have obtained the rms torque,
, on a mass shell using Eq. (6) then we obtained
the total specific angular momentum, , acquired
during expansion integrating the torque over time (R88a Eq. 35):
![[EQUATION]](img59.gif)
where the functions ,
are given by R88a (Eq. 31) while the mean over-density inside the
shell, , is given by (R88a):
![[EQUATION]](img63.gif)
In Fig. 1 we show the variation of with the
distance r for three values of the peak height
. The rms specific angular momentum,
, increases with distance r while peaks
of greater acquire less angular momentum via
tidal torques.
![[FIGURE]](img70.gif) |
Fig. 1. The specific angular momentum, in units of , Mpc and the Hubble time, , for three values of the parameter ( dotted line, solid line, dashed line) and for .
|
This is the angular momentum-density anticorrelation showed by
Hoffman (1986). This effect arises because the angular momentum is
proportional to the gain at turn around time, ,
which in turn is proportional to .
© European Southern Observatory (ESO) 1998
Online publication: August 6, 1998
helpdesk.link@springer.de  |