X-Ray Studies of
Nucleosynthesis and
Abundances in Supernova
Remnants
John P. Hughes
Rutgers University
February 20, 2003
Carnegie Symposum
1
Why X-rays?


Lya lines of all species from C (0.368 keV) to Zn (9.3 keV)
kT ~ 106 K to 108 K from shocks in ejecta and CSM/ISM
ASCA
Si
W49B
S
Ar
Ca
Fe
February 20, 2003
Carnegie Symposum
2
X-ray Emission/Atomic Processes
Continuum emission – thermal bremsstrahlung:
Line emission:
Abundance
of element Z
February 20, 2003
Ionization fraction
of ion i
Carnegie Symposum
3
Abundance Determination Issues

Thermodynamic State
– Nonequilibrium Ionization (NEI) (net~105 cm-3 yr)
– T, n evolution with time/radius (e.g., Sedov)
– Other effects:
 Heating/cooling in pure element ejecta
 Te/Tp
 Nonthermal particles (rates and excitation)

Absolute abundances (e.g., Si/H, O/H, Fe/H)
– Rely on assumption of H/He-dominated continuum

Relative abundances (e.g., Mg/Si, O/Fe)
– OK, if species have the same spatial distribution
February 20, 2003
Carnegie Symposum
4
Ejecta Mass Determination Issues
Volume estimation
 Clumping (reduces actual mass)
 Distance (M~D5/2)
 Source of electrons

–
–
–
–
Measure EM = nenIV
Solar abundance: ne ~ nH ~ nFe/107.6-12 ~ 25000nFe
Pure Fe: ne ~ 20nFe
Inferred Mpure Fe /Msolar ~ 35
February 20, 2003
Carnegie Symposum
5
Where do we find ejecta?
Optical: SNRs with high velocity oxygen-rich features
Galactic: Cas A, G292.0+1.8, Puppis A
LMC/SMC: N132D, E0540-69.3, E0102.2-72.2
Other: an unresolved SNR in NGC 4449
 Remnants of historical SNe
e.g., SN1006, SN1572 (Tycho), SN1604 (Kepler)
Based on [Fe II] in absorption; X-ray spectra
 Ejecta-dominated SNRs
e.g., W49B, G352.7-0.1, G337.2-0.7, G309.2-0.6
Based on X-ray spectra (mostly ASCA)
 Nearly all remnants up to ages of at least ~10,000 yrs!!!
N49, N63A, DEM71, N49B, and E0103-72.6
Based on Chandra spectro-imaging

February 20, 2003
Carnegie Symposum
6
Core Collapse Supernovae

SN II, SN Ib/c (Zwicky & Baade 1934)
– Massive stars that explode with (SN II) or
w/out (SN Ib/c) their H envelopes
– Photodisintegration of Fe, plus electron
capture on nuclei, remove central P support
– Core collapses, leading to NS or BH
– Core stiffens, rebounds and drives an outward
moving shock
– Neutrinos or jets needed to produce explosion
– Mean Rate ~ 1.3 SNU
February 20, 2003
Carnegie Symposum
7
Nucleosynthesis in CC SNe

Hydrostatic nucleosynthesis
– During hydrostatic evolution of star
– Builds up shells rich in H, He, C, O, and Si
– Amount of C, O, Ne, Mg ejected varies strongly with
progenitor mass

Explosive nucleosynthesis
– Some mechanism drives a shock wave with 1051+ erg
through the Fe-core
– Burning front T’s of ~109 K cause explosive O- and Siburning
– Only affects the central parts of the star – outer
layers retain their pre-SN composition
February 20, 2003
Carnegie Symposum
8
Explosive Nucleosynthesis
Process
T (109 K)
Main Products
Explosive complete Si-burning
5.0
“Fe”, He
Explosive incomplete Si-burning
4.0
Si, S, Fe, Ar, Ca
Explosive O-burning
3.3
Explosive Ne/C-burning
February 20, 2003
1.2
Carnegie Symposum
O, Si, S, Ar, Ca
O, Mg, Si, Ne
9
Overturning Our View of Cas A
Hughes, Rakowski, Burrows, and Slane 2000,
ApJL, 528, L109.
February 20, 2003
Carnegie Symposum
10
Cas A - Doppler Imaging by XMM

Similar velocity structures in
different lines
– SE knots blueshifted
– N knots redshifted
– Tight correlation between Si
and S velocities

Fe
– Note velocity distribution in N
– Extends to more positive
velocities than Si or S
Willingale et al 2002, A&A, 381, 1039
February 20, 2003
Carnegie Symposum
11
Cas A – 3D Ejecta Model
“Plane of the sky”
“Rotated”
Red: Si Ka
Green: S Ka
Blue: Fe Ka
Circle: Main shock
Fe-rich ejecta lies outside Si/S-rich ejecta
February 20, 2003
Carnegie Symposum
12
Oxygen-Rich SNR G292.0+1.8
Park et al 2001, ApJL, 564, L39
February 20, 2003
Carnegie Symposum
13
Oxygen-Rich SNR G292.0+1.8
Ejecta
Rich in O, Ne, and Mg, some Si
[O]/[Ne] < 1
No Si-rich or Fe-rich ejecta
February 20, 2003
Carnegie Symposum
14
Oxygen-Rich SNR G292.0+1.8
Normal Composition, CSM
Central bright bar – an axisymmetric
stellar wind (Blondin et al 1996)
Thin, circumferential filaments enclose
ejecta-dominated material – red/blue
supergiant wind boundary
February 20, 2003
Carnegie Symposum
15
Thermonuclear Supernovae

SN Ia (Hoyle & Fowler 1960)
– No hydrogen, a solar mass of 56Ni, some
intermediate mass elements (O, Mg, Si, S,…)
– Subsonic burning (deflagration) of approx.
one Chandrasekhar mass of degenerate C/O
– C-O white dwarf accreting H/He-rich gas from
a companion
– No compact remnant
– Mean rate ~ 0.3 SNU
February 20, 2003
Carnegie Symposum
16
Identifying Remnants of SN Ia

Tycho


Balmer-dominated SNRs
(partially neutral ISM)
Ejecta abundances (Si and Fe
rich, poor in O and Ne)
Remnant structure (uniform
ISM, “smoother” ejecta, little
spectral variation)
E0509-67.5
February 20, 2003
Carnegie Symposum
17
SN Ia Spectra and Abundances
 Comparison to models
 O, Ne, Mg: relatively low
 Si, S, Ar, Ca: consistent
 Fe: very low (<0.1)
 Other spectral results
 Fe: co-spatial with Si, but
hotter and lower net
W7: Nomoto et al 1984, Thielemann et al 1993
WDD1: Iwamoto et al 1999
NEI fit: Warren et al 2003
February 20, 2003
Carnegie Symposum
18
ISM Abundances of the LMC
Using SNRs as a probe of the ISM
 7 SNRs, ages from 2,000 yr to 20,000 yr
 Data from ASCA
 Spectra calculated for evolutionary models
(Sedov solution)

– spatial variation
– temporal variation
February 20, 2003
Carnegie Symposum
19
LMC SNRs: Integrated Abundances
From fits to ASCA global X-ray spectra of 7 evolved LMC remnants
N49B
DEM L71
Hughes, Hayashi, & Koyama 1998, ApJ, 505, 732
February 20, 2003
Carnegie Symposum
20
LMC Metal Abundances
Species
HHK98
Duf84
RD92
O
8.21(7)
8.43(8)
8.35(6)
Ne
7.55(8)
7.64(10)
7.61(5)
Mg
7.08(7)
...
7.47(13)
Si
7.04(8)
...
~7.8
S
6.77(13)
6.85(11)
6.70(9)
Fe
7.01(11)
...
7.23(14)
HHK95: ASCA X-ray SNRs
Duf84: UV/Optical spectra H II regions (Dufour 1984)
RD92: F supergiants (Mg, Si, Fe) (Russell & Bessel 1989)
H II regions, SNRs (O, Ne, S) (Russell & Dopita 1990)
February 20, 2003
Carnegie Symposum
21
DEM L71

Middle-aged SNR
– 36” (8.7 pc) in radius
– 4,000 yrs old



Rims: LMC composition
Core: [Fe]/[O] > 5 times solar
Ejecta mass: 1.5 Msun
SN Ia ejecta
Hughes, Ghavamian, Rakowski, & Slane 2003, ApJ, 582, L95
February 20, 2003
Carnegie Symposum
22
N49B

Middle-aged SNR
– 80” (19 pc) in radius
– 5000-10,000 yrs old

Bright and faint rims
– LMC composition
– ISM density varies by x10

Ejecta
– Revealed by equivalent-width maps
– Mg & Si rich, no strong O or Ne
Park, Hughes, Slane, Burrows, Garmire, & Nousek 2003,
ApJ, submitted.
February 20, 2003
Carnegie Symposum
23
SNR 0103-72.6

Middle-aged SNR
– 87” (25 pc) in radius
– >10,000 yrs old (?)

Circular rim
– SMC composition

Central bright region
– O, Ne, Mg, Si-rich ejecta
– No Fe enhancement
Park, et al 2003, ApJ, in prep.
February 20, 2003
Carnegie Symposum
24
Summary

Core Collapse SNe
– Cas A
 X-ray ejecta dominated by Si, S, and Fe
– explosive nucleosynthesis
 Extensive mixing and overturning of ejecta layers
– G292.0+1.8
 X-ray ejecta dominated by O, Ne, and Mg (no Fe)
 Ambient medium strongly modified by progenitor
 Contains “normal” young pulsar and its wind nebula
Highly Structured Ejecta/Environment
February 20, 2003
Carnegie Symposum
25
Summary

Thermonuclear SNe (Tycho, E0509-67.5)
– X-ray ejecta dominated by Si, S, and Fe
– Stratification (most of the Fe core unshocked)
– Fe: higher kT, lower net – due to:
 evolution (ejecta density profile)
 radioactivity
– Ejecta relatively smooth and symmetric
 only factor of 2 intensity variations
 little spectral variation
 few (one or two) clumps of Fe-rich ejecta
February 20, 2003
Carnegie Symposum
26
Summary

Evolved LMC SNRs
– Global X-ray abundances consistent with
optical/UV values
– Individual SNRs show obvious signs of ejecta
 DEM L71: 4,000 yrs, Si and Fe-rich SN Ia ejecta
 N49B: 5,000-10,000 yrs, Mg and Si-rich ejecta
 E0103-72.6: 10,000+ yrs, O, Ne, and Mg-rich ejecta

Issues for nucleosynthesis models
– O/Ne ratio < 1 (G292.0+1.8)
– O,Ne/Mg << 1 (N49B)
February 20, 2003
Carnegie Symposum
27
THE END
February 20, 2003
Carnegie Symposum
28
SN Ia Spectra and Abundances
February 20, 2003
Carnegie Symposum
29
Properties of DEM L71 Ejecta




Outer rim: lowered abundances, ~0.2 solar (LMC ISM)
Core: enhanced Fe abundance, [Fe]/[O] > 5 times solar (ejecta)
Thick elliptical shell, 32” by 40” across (3.9 pc by 4.8 pc)
Dynamical mass estimate
Wang & Chevalier 2001
r’ ~ 3.0
Mej = 1.1 Mch (n/0.5 cm-3)

EM mass estimate
EM ~ ne nFe V
MFe < 2 Msun

Main error: source of electrons
Fe-rich, low mass
February 20, 2003
SN Ia
Carnegie Symposum
30
N63A

Middle-aged SNR
– 34” (8.2 pc) in radius
– 2000-5000 yrs old
– 2nd brightest LMC SNR

“Crescent”-shaped features
– Similar to features in Vela
– Clumps of high speed ejecta
– Not ejecta dominated

Triangular hole
– X-ray absorption
– Approx. 450 solar mass cloud
– On near side

No PSR or PWN
– LX < 4x1034 erg s-1
Warren, Hughes, & Slane, ApJ, in press (20 Jan 2003)
February 20, 2003
Carnegie Symposum
31