Improved Absorption Models For The ISM: Jörn Wilms Dr. Remeis-Sternwarte & Ecap

Improved Absorption Models for the ISM
Jrn Wilms
Dr. Remeis-Sternwarte &
ECAP
in collaboration with:
J.C. Lee (Harvard), M.A. Nowak (MIT), N.S. Schulz (MIT), A. Juett (Ex-GSFC)
Black Hole Universe
Photoabsorption in ISM: strong

modification of observed spectrum
= To understand source physics,
we need to understand this
modification and correct for it
as precisely as possible
Observed Flux [erg cm2 s1 keV1]
31019
11020
31020
11021
= X-ray astronomers should really care about the subject of

this talk.
31021
11022
31022
0.1
1.0
Energy [keV]
10.0
= How many of us care about absorption in the ISM?

ADS: Total number of papers from 20002010 using XMM-Newton, Chandra, or
Suzaku:
5700

Suzaku:
5700
Total number of 20002010 papers citing an absorption paper:
1330

Suzaku:
5700
1330
Only 23% of all X-ray astronomers care enough about absorption in the ISM
to mention it in their papers!

Suzaku:
5700
1330
Of this:
Morrison & McCammon (1983): 767 (58%)
Baucinska-Church
& McCammon (1992): 247 (19%)
Wilms et al. (2000): 316 (24%)

Suzaku:
5700
1330
Of this:
Morrison & McCammon (1983): 767 (58%)
Baucinska-Church
& McCammon (1992): 247 (19%)
Wilms et al. (2000): 316 (24%)
Only 43% of all citations, and 10% of all X-ray papers cite a modern absorption paper!
Typical life of an X-ray astronomer:
solaris:~/data> make proposal

linux:~/data> make observation

macos:~/data> make datareduction

macos:~/data> xspec
Xspec 15.3.1xy 21:14:59 05-Mar-2020
For documentation, notes, and fixes see
http://xspec.gsfc.nasa.gov/
Plot device not set, use "cpd" to set it
Type "help" or "?" for further information
XSPEC>data coolobs.pha

macos:~/data> xspec
Xspec 15.3.1xy 21:14:59 05-Mar-2020
For documentation, notes, and fixes see
http://xspec.gsfc.nasa.gov/
Plot device not set, use "cpd" to set it
Type "help" or "?" for further information
XSPEC>data coolobs.pha
XSPEC>model wabs*power
= and this is where a mistake is made!
also promoted until recently in xspec manual thanks for changing this!
Observed Flux [erg cm2 s1 keV1]
31019
Photoabsorption in ISM: modification of observed spectrum

= To understand source physics,
we need to understand this
modification and correct for it
11020
31020
11021
Here, I will discuss three subjects

in more detail:
1. Elemental abundances
2. Photoabsorption cross section
3. Influence of dust
31021
11022
31022
Any errors in the above will distort the inferred souce spectrum!
0.1
1.0
Energy [keV]
10.0
0.1
1.0
10.0
Ni
700
Fe
E3keV [1024 cm2]
600
400
300
[Mbarn/H]
101
102
103
104
105
106
0.1
1.0
E [keV]
S Ar
10.0
Ca
Cr
Si
Mg Al
200
Ne
FeL
100
N O
H+He
H2
0.1
1.0
10.0
E [keV]
Although you use NH to describe the column, what you measure in X-rays is
the column of metals.
= Knowing abundances is important to be able to compare NH measurements

from multiwavelength observations.
12
Anders & Grevesse (1989)
Lodders (2003)
Asplund et al. (2009)
Wilms et al. (1999)
log Abundance (H=12)
10
2
H
He
Li
Be
Na Al
P
Cl
K
Sc
V Mn Co Cu
Ne Mg Si
S
Ar Ca Ti
Cr Fe Ni
XSPEC-Default: Solar abundance of Anders & Grevesse (1989), which is outdated.

Best current values for Sun: Asplund et al. (2009).
Abundance/(Wilms et al., 1999)
2.0

Lodders (2003)
1.8
1.6
1.4
1.2
1.0
0.8
H
He
Li
Be
Na Al
P
Cl
K
Sc
V Mn Co Cu
Ne Mg Si
S
Ar Ca Ti
Cr Fe Ni
Solar abundances of Anders & Grevesse (1989): 40% higher than ISM
and 20% higher than modern solar abundance.
Therefore use Wilms et al. (2000) abundances. Note that Sun is still overabundant wrt. ISM!

Lodders (2003)
(abundance set)/(Wilms et al., 1999)
1.8
1.6
1.4
1.2
1.0
0.8
2
10
Energy [keV]
To obtain proper NH, use Wilms et al. (2000), if not, your column will be
10%20% too small.
abund wilm in XSPEC. . . ,
generally gives consistent results with other measurements, cf. Weisskopf et al. (2004).
2nd and 3rd subject: Cross sections and solid state effects.
1.10
Fe cross section
BCMC/Vern
1.05
1.00
0.95
0.90
0.1
1.0
Energy [keV]
10.0
Three sets of cross sections available in fitting programs:

wabs: Total cross section only (no free abundances)
xsect bcmc: Fits to Henke et al. (1993) by Baucinska-Church

& McCammon (1992)
xsect vern: Fits to ab initio QM calculations by Verner et al. (1996) (most trustworthy)
In addition: solid state effects:
Self-shielding: dust reduces absorptivity by shielding = Wilms et al. (2000)
Modification to cross sections around edges (XAFS)
see Lee & Ravel (2005) and Lee et al. (2009) for introductions for astrophysicists
Over the past years we have updated the code of Wilms et al. (2000):
tbnew
http://pulsar.sternwarte.uni-erlangen.de/wilms/research/tbnew
Main advantages:
significantly faster than phabs, tbabs, comparable to wabs
Due to internal caching algorithms
tbnew
Main advantages:
can be used to fit relative abundances or absolute columns

isis> list_free;
idx param
tie-to freeze value
1 tbnew(1).nH 0
0
1.2
6 tbnew(1).Ne 0
0
0.591
min
0
0
max
10 1022
5
vs.
isis> list_free;
idx param
tie-to freeze value
min
6 tbnew(1).Ne 0
0 -8.7e17 -1e19
max
0
tbnew
Main advantages:
10.00
0.5 1 2
O I 1s2p
O II
O III
10
20
bf [Mbarn]
1.00
0.10
1 0 1 0
Counts (per bin)
Flux (105 f.u.)

includes the detailed structure of important edges (Fe-L, O-K, Ne-K, [Mg-K])
0.01
864
22
24
Wavelength []
866
868
Energy [eV]
870
872
26
O-edge theory: Gorczyca & McLaughlin (2000)

Observation: Cygnus X-1 (Hanke et al., 2009)
Ne-edge: Verner et al. (1996) vs. Gorczyca (2000)

not trivial: need smooth transition to Verner et al.
(1996)!
tbnew
Main advantages:
10
Fe L3
10
bf [Mbarn]
20
Fe L2
0
2
Counts (per bin)
Flux (105 f.u.)

includes the detailed structure of important edges (Fe-L, O-K, Ne-K, [Mg-K])
started to include solid-state effects (cf. Lee & Ravel, 2005; Lee et al., 2009)
17
17.5
Wavelength []
Fe L-edge in Cygnus X-1 (Hanke et al., 2009)
18
700
710
720
Energy [eV]
730
740
Fe L-edge: Verner et al. (1996), Brennan & Cowan

(1992), Kortright & Kim (2000)
17.46
Wavelength (104 m)
17.22
16.98
17.71
0.4
Transmission (eN)
0.6
0.8
Further extension to tbnew: bcdust (or

similar name)
Step 1: Interface to external cross section data
FITS-files listing energy and cross
section
Specification of elemental composition to allow depletion calculation
Fe LII
Continuum absorption by FeL
Fe metal
Fayalite (Fe2SiO4)
Hematite (Fe2O3)
Lepidocrocite FeO(OH)
Iron Sulfate (FeSO4)
Molecules/Solids Fe LIII Fe metal measured at ALS

(Lee et al. 2009)
(Kortright & Kim 2000)
700
710
720
Energy (eV)
Optional extrapolation (MBACK/splines) to get smooth transition to vacuum

cross section
. . . code done in IDL; rewrite in C almost done, after this: testing phase; release 2012 Oct./Nov.
If this works:
build library of compound cross sections
see http://xafs.org/JuliaLee/Standards and Lee et al. (2009)
730
Hanke (2010, PhD Thesis
Step 2: halo modeling?

Plans so far. . .
Using assumed dust composition and cross sections, build halo model
Interface with point spread function to allow proper fitting of point sources with
halo contribution
Difficult to do in XSPEC data model, but can do this in ISIS and (perhaps) Sherpa
One of the most important discoveries by INTEGRAL: IGR sources, i.e., population of accreting neutron stars with NH 1022 cm2.
(Walter et al., 2004; Filliatre & Chaty, 2004)
up to 50% of all neutron stars in
102
Galaxy
Fluorescent Fe K line can contribute >50% of flux <10 keV
Fe line: ionization state of wind
normalized counts s
cm 2
101
(ratio Fe K/Fe K not well known!)
Absorber geometry from Compton shoulder

103
(e.g., Matt & Guainazzi, 2003; Watanabe

et al., 2003)
1
6
10
Energy [keV]
40
(Barragn et al., 2009; Barragn et al., 2011)
0.01
105
Photons cm2 s1 keV1

104
103
10
20
Energy (keV)
IGR J163184848: Model (100 ksec)
50
1
0.01
Counts s1 keV1
0.1
5
10
Energy (keV)
IGR J163184848: Fe-Band (100 ksec)
0.
300
[lt-sec]
400
Fe K edge
Fe K beta
Ni XXVII
Counts/s/Angstrom
K
Ca
500
200
0.
0.7
100
0
0.01
ss
e
5. c 1
]
3. 00
75
2.
5
1. 0
0. 25
00
0.2
SK
Ar
K
0.3
0.1
Fe Kalpha
0.4
[ct
600
700
0.
Cts sec1 keV1
10
Cr K
Ni
Fe K
Fe
K
K
a)
Ni
100
0.8
0.
-100
-2
-4
4 c)
2
2
2.2
Wavelength [Angstrom]
2.4
9
0.
20
-400
XMM
0555200401
To Earth
-500
-600
-200
0
[lt-sec]
200
Probably low ionization stages, but difficult to measure with

XMM-Newton.
-2
-4
4 f)
2
0
-2
-4
2
-400
Pre-periastron flare of Be-system GX 3012:

GX 3012: Porb = 41.5 d, Ppulse 685 s, e = 0.47
NH 1024 cm2
large number of flourescent lines
-2
-4
4 e)
2
-300
1.8
-2
-4
4 d)
2
-200
10-3
4 b)
2
6
Energy [keV]
10
(Frst et al., 2011, Torrejn et al., 2011)
400
10
0.1
keV
0.01
3
Counts s
10
Photons cm 2 s 1keV 1
10
5
10
10
Energy (keV)
20
10
Energy (keV)
GX 3012: Model and soft lines (100 ksec; assuming broadish lines)
= Astro-H will resolve Compton shoulder, separate Fe K and Fe edge
Eikmann et al. (2012, in prep)
Further improvement of absorption code: optically thick absorption

requires convolution model
Fluorescence lines
Eikmann et al. (2012, in prep)
Further improvement of absorption code: optically thick absorption

requires convolution model
Fluorescence lines
Compton broadening
default
32a
Bibliography
Anders E., Grevesse N., 1989, Geochim. Cosmochim. Acta 53, 197
Asplund M., Grevesse N., Sauval A.J., Scott P., 2009, Ann. Rev. Astron. Astrophys. 47, 481
Baucinska-Church
M., McCammon D., 1992, Astrophys. J. 400, 699
Barragn L., Wilms J., Kreykenbohm I., et al., 2011, Proceedings of Science INTEGRAL 2010, 135
Barragn L., Wilms J., Pottschmidt K., et al., 2009, Astron. Astrophys. 508, 1275
Brennan S., Cowan P.L., 1992, Rev. Sci. Instrum. 63, 850
Filliatre P., Chaty S., 2004, Astrophys. J. 616, 469
Frst F., Suchy S., Kreykenbohm I., et al., 2011, Astron. Astrophys. 535, A9
Gorczyca T.W., 2000, Phys. Rev. A 61, 024702
Gorczyca T.W., McLaughlin B.M., 2000, J. Phys. B 33, 2000
Hanke M., Wilms J., Nowak M.A., et al., 2009, Astrophys. J. 690, 330
Henke B.L., Gullikson E.M., Davis J.C., 1993, Atomic Data Nucl. Data Tables 54, 181 available at http://cindy.lbl.gov/optical_constants/
Kortright J.B., Kim S., 2000, Phys. Rev. B 12216
Lee J.C., Ravel B., 2005, Astrophys. J. 622, 970
Lee J.C., Xiang J., Ravel B., et al., 2009, Astrophys. J. 702, 970
Matt G., Guainazzi M., 2003, Mon. Not. R. Astron. Soc. 341, L13
Morrison R., McCammon D., 1983, Astrophys. J. 270, 119
Verner D.A., Ferland G.J., Korista K.T., Yakovlev D.G., 1996, Astrophys. J. 465, 487
Walter R., Courvoisier T.J.L., Foschini L., et al., 2004, In: Proc. 5th INTEGRAL Workshop: The INTEGRAL Universe. ESA SP-552, ESA, Noordwijk, p. 417
0
Extension to large NH
32b
Watanabe S., Sako M., Ishida M., et al., 2003, Astrophys. J. 597, L37
Weisskopf M.C., ODell S.L., Paerels F., et al., 2004, Astrophys. J. 601, 1050
Wilms J., Allen A., McCray R., 2000, Astrophys. J. 542, 914

Improved Absorption Models For The ISM: Jörn Wilms Dr. Remeis-Sternwarte & Ecap

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Improved Absorption Models For The ISM: Jörn Wilms Dr. Remeis-Sternwarte & Ecap

Uploaded by

Copyright:

Available Formats

Improved Absorption Models for the ISM

Black Hole Universe

Photoabsorption in ISM: strong

Observed Flux [erg cm2 s1 keV1]

= X-ray astronomers should really care about the subject of

= How many of us care about absorption in the ISM?

= How many of us care about absorption in the ISM?

= How many of us care about absorption in the ISM?

= How many of us care about absorption in the ISM?

= How many of us care about absorption in the ISM?

= How many of us care about absorption in the ISM?

Typical life of an X-ray astronomer:

solaris:~/data> make proposal

Typical life of an X-ray astronomer:

solaris:~/data> make proposal

Typical life of an X-ray astronomer:

solaris:~/data> make proposal

Typical life of an X-ray astronomer:

solaris:~/data> make proposal

Typical life of an X-ray astronomer:

solaris:~/data> make proposal

Observed Flux [erg cm2 s1 keV1]

Photoabsorption in ISM: modification of observed spectrum

Here, I will discuss three subjects

E3keV [1024 cm2]

= Knowing abundances is important to be able to compare NH measurements

log Abundance (H=12)

XSPEC-Default: Solar abundance of Anders & Grevesse (1989), which is outdated.

Abundance/(Wilms et al., 1999)

Anders & Grevesse (1989)

Anders & Grevesse (1989)

(abundance set)/(Wilms et al., 1999)

Three sets of cross sections available in fitting programs:

xsect bcmc: Fits to Henke et al. (1993) by Baucinska-Church

can be used to fit relative abundances or absolute columns

Counts (per bin)

Flux (105 f.u.)

can be used to fit relative abundances or absolute columns

O-edge theory: Gorczyca & McLaughlin (2000)

Ne-edge: Verner et al. (1996) vs. Gorczyca (2000)

Counts (per bin)

Flux (105 f.u.)

can be used to fit relative abundances or absolute columns

Fe L-edge in Cygnus X-1 (Hanke et al., 2009)

Fe L-edge: Verner et al. (1996), Brennan & Cowan

Further extension to tbnew: bcdust (or

Molecules/Solids Fe LIII Fe metal measured at ALS

Optional extrapolation (MBACK/splines) to get smooth transition to vacuum

Hanke (2010, PhD Thesis

Step 2: halo modeling?

up to 50% of all neutron stars in

(ratio Fe K/Fe K not well known!)

Absorber geometry from Compton shoulder

(e.g., Matt & Guainazzi, 2003; Watanabe

(Barragn et al., 2009; Barragn et al., 2011)

Photons cm2 s1 keV1

IGR J163184848: Model (100 ksec)

IGR J163184848: Fe-Band (100 ksec)

Cts sec1 keV1

Probably low ionization stages, but difficult to measure with

Pre-periastron flare of Be-system GX 3012:

(Frst et al., 2011, Torrejn et al., 2011)

Eikmann et al. (2012, in prep)