![]() | ![]() |
Astron. Astrophys. 351, 759-765 (1999) 3. Attenuation of X-rays and
|
![]() |
Fig. 2. The combined photoelectric and Compton cross-sections relative to H as a function of energy, scaled by E/10 keV for clarity of presentation. (a) The elements H to Fe that are at least as abundant as Fe (H, He, C, N, O, Ne, Mg, Si, Fe) (b) the same elements as in (a) with solar abundance of H and He reduced by 10 (c) The precursor dust combination ![]() |
The elements O, Si, Mg and Fe dominate the composition of the
chondrules but the composition of the precursor grains has been the
subject of much study and speculation (Hewins, 1997). A number of
chondrule classification systems have been adopted and early
approaches depended mainly on either bulk composition or texture or
both. McSween (1977) recognised two main types, type I or FeO poor and
type II or FeO rich. However in comparison with chondrules, the
fine-ground matrix in chondritic meteorites is more FeO rich than even
type II chondrules and it has been proposed that this matrix may be
close to the composition of the chondrule precursor (Huang et al.,
1996). In this scenario, the precursors of type I chondrules were
enriched in Fe and more efficiently heated by x-ray and
-ray absorption, resulting in the loss
of Fe and other volatile elements that ended in the enriched matrix.
Type II chondrules appear not to have lost significant amounts of
volatiles in the melting process. The composition of the precursor
grains may be resolved by new x-ray and
-ray heating experiments. A
composition consisting of solar abundance of oxides of Fe, Si and Mg
or
was assumed for the precursor
grains and the product of the cross-section of this combination by
abundance relative to H is plotted in Fig. 2c. The x-ray and
-ray absorption efficiencies of
different thicknesses of precursor grains, assuming a density of one,
are given in Fig. 3 and grains in the range 10 µm to 1 cm
are very efficient absorbers in the region where dust dominates the
absorption (Fig. 2). This range agrees quite well with the measured
Weibull and lognormal distributions of chondrule sizes (Martin &
Hughes, 1980; Rubin & Keil, 1984). The deficiency of small grains
is caused by low absorption efficiency and substantial radiation
losses from grains with large surface to volume ratios.
![]() |
Fig. 3. The absorption efficiency of different thickness of ![]() |
The thickness of the dust layer converted to chondrules depends on
the GRB spectrum which must have significant emission below 30 keV
where dust absorption dominates (Fig. 2) and also on the mixture and
distribution of gas and dust in the nebula. For a GRB with
1053 ergs and an assumed spectrum
= -1,
= - 2 and
= 15 keV (Fig. 1), the fraction of
GRB energy photoelectrically absorbed by the dust is 20%, increasing
to 27% for a factor 10 reduction in H and He. In the simplified case
of solar abundance and a uniform mix of gas and dust, the thickness of
the chondrule layer created is 0.18 g cm-2 corresponding to
one optical depth for 30 keV x-rays. The layer thickness increases to
about 0.8 g cm-2 and 2.0 g cm-2 for optical
depths to 40 keV and 55 keV x-rays with H and He abundances reduced by
factors of 3 and 10 respectively. The thickness of the chondrule layer
is therefore controlled by the degree of gas depletion from the
nebula. The minimum GRB fluence required to produce chondrule layers
of 0.18, 0.8 and 2.0 g cm-2 is 1.8
1010, 7.0
1010 and 1.5
1011 ergs cm-2, adopting 20%, 23% and 27%
absorption by the chondrule precursors and 2
1010 erg g-1
for heating and melting. A fluence of
1011 erg cm-2 implies a distance of about 100 pc
to the source for an output of 1053 ergs radiated
isotropically. The GRB would also form a layer of chondrules over a
large area (103 - 104 pc2) in a
nearby molecular cloud provided large precursor grains had already
formed (Weidenschilling & Ruzmaikina, 1994). The process of
chondrule amalgamation might be sufficient to trigger star formation
over this region. In this case chondrule formation precedes cloud
collapse and star formation. The existence of pre-solar grains in
meteorites is well established (Zinner, 1996) but there is no evidence
for pre-solar chondrules.
The chondrules cooled at a much slower rate than if they were isolated (Hewins, 1997). They may have been warmed by a fading source or by forming a thermal blanket or a combination of both effects. BeppoSAX discovered x-ray afterglow from GRB sources and the limited measurements show considerable variability between the various GRBs (Costa et al., 1997; Piro et al., 1998). The afterglow typically decreases by at least a factor of 20 in 103 s which yields more than a factor of two drop in temperature. This decrease of 3800 K hr-1 is too rapid to account for the chemical and textural properties of chondrules (Yu & Hewins, 1998). The chondrules on the far side of the layer from the GRB source cool even more rapidly because of shielding by foreground chondrules and spectral softening of the afterglow. The optical depth of the chondrules to their infrared radiation at a peak of about 1.5 µm is about 0.25 g cm-2 assuming all the chondrules have size 0.1 cm (Hood & Horanyi, 1991; Wood, 1988). The x-ray and infrared optical depths are comparable and the cooling rate was further reduced by this thermal blanket.
There are several indicators that the dust was concentrated and/or
gas depleted in the nebula when chondrules formed. These include: (1)
the increased rate of collisions between plastic and molten chondrules
to form adhering pairs (Wasson, 1993), (2) the seeding of melted
chondrules with dust grains (Connolly & Hewins, 1995), (3) the O/H
ratio well above the solar value (Fegley & Palme, 1985), and (4)
the improved absorption efficiency of x-rays and
-rays by the precursor dust balls. The
rims on chondrules indicate time spent in dusty regions.
© European Southern Observatory (ESO) 1999
Online publication: November 3, 1999
helpdesk@link.springer.de