Chandra Observations of Supernova 1987A

1
Chandra Observations of Supernova 1987A
SN 1987A 20 Years after SNe and GRBs, Aspen,
February 19, 2007
Sangwook Park
Department of Astronomy and Astrophysics
The Pennsylvania State University
With David Burrows (Penn
State) Judith Racusin (Penn State) Gordon Garmire
(Penn State) Richard McCray (Colorado) Svetozar
Zhekov (Colorado, SRI)
2
SN 1987A in Her Uniqueness

• Brightest supernova observed by mankind since
  1604 (J. Kepler)
• Distance 50 kpc, in the LMC
• Age 20 years old as of Feb 2007
• Type II SN
• Progenitor Blue supergiant (Sk -69 202, B3 I)
• Neutrino burst
• Core-collapse explosion neutron star?
• Most intensively studied SN of all time
• Optical/UV HST and many ground-based
• Radio initial detection, turned on again in
  1990
• X-ray no initial detection, turned on in 1990
• Gamma-ray detected decay lines from 56Co ? decay
  of 56Ni,
• confirming explosive nucleosynthesis
• ADS 1000 (1/week) refereed papers (since
  1987)

• Chandra monitoring since 1999
• Twice a year, separated by 6 months
• As of 2007-01, 16 observations performed

3
SN 1987A Chandra Observations

• Date Instruments Exp.
  
• (since SN)
  (ks)
• 1999-10-6 (4609) ACIS-SHETG 116
• 2000-1-17 (4711) ACIS-S3 9
• 2000-12-7 (5038) ACIS-S3 99
• 2001-4-25 (5176) ACIS-S3 18
• 2001-12-12 (5407) ACIS-S3 49
• 2002-5-15 (5561) ACIS-S3 44
• 2002-12-31 (5791) ACIS-S3 49
• 2003-7-8 (5980) ACIS-S3 45
• 2004-1-2 (6157) ACIS-S3 46
• 2004-7-22 (6359) ACIS-S3 49
• 2004-8-26 (6393) ACIS-SLETG 289
• 2004-9-5 (6404)
• 2005-1-9/13 (6533) ACIS-S3 48
• 2005-7-11/16 (6716) ACIS-S3 44
• 2006-1-28 (6914) ACIS-S3 42
• 2006-7-27 (7095) ACIS-S3 40
• 2007-1-19 (7271) ACIS-S3 38

- Publications Burrows et al. 2000, ApJ, 543,
L149 (Obs 1-2) Park et al. 2002, ApJ, 567, 314
(Obs 1-4) Michael et al. 2002, 574, 166 (Obs 1
3) Park et al. 2004, AdSpR, 33, 386 (Obs
1-6) Park et al. 2004, ApJ, 610, 275 (Obs
1-7) Park et al. 2005, AdSpR, 35, 991 (Obs
1-9) Zhekov et al. 2005, ApJ, 628, L127 (Obs
11) Park et al. 2005, ApJ, 634, L73 (Obs 2-10,
12-13) Zhekov et al. 2006, ApJ, 645, 293 (Obs
2-10, 12-13) Park et al. 2006, ApJ, 646, 1001
(Obs 2-10, 12-13) Racusin et al. 2007, in
preparation HEAD 2006-10 (X-ray radial
expansion Obs 1-15)
Current presentation A review and the latest
results.
4
SN 1987A Physical Picture
Optical/Soft X-rays
Hard X-rays
Radio
?
NS/BH
Artistic presentation of SN 1987A (SAO/CXC)
Cf. Michael et al. 1998
5
SN 1987A First X-ray Images
N
E
ACIS (1999-10) Burrows et al. 2000
ROSAT/HRI (5 pixels) HEASARC/SkyView
Green-Blue ACIS Red HST Contour ATCA
1 arcsecond
6
SN 1987A ACIS Images 20002007
N
Ring-like, asymmetric intensity Developments of
X-ray spots becoming a complete ring as the
blast wave arrives the inner ring! Surface
brightness increase Now 18 x brighter than
2000 Lx (0.5-2keV) 2.1 x 1036 ergs/s No
point source at the center
E
2001-04-25
2000-12-07
2003-07-08
2002-12-31
Scheduled on 2007-7-16
2005-07-11
2005-01-09
1 arcsec
7
SN 1987A ACIS Subband Images (2000-2004)
2000-12
2001-12
2002-12
2004-01
0.3-0.8 keV
(HST 2000-11)
(HST 2001-12)
(HST 2001-12)
(HST 2001-12)
0.8-1.2 keV
1.2-8.0 keV
(ATCA 1999-9)
(ATCA 2001-11)
(ATCA 2002-11)
(ATCA 2002-11)
1 arcsec
8
SN 1987A Soft X-Ray Intensity Ratio
2002-12 to 2000-12
2005-7 to 2002-12
(5791) (5038)
(6716) (5791)
(0.5 2 keV)
(0.5 2 keV)
Contours 2002-12
Contours 2005-7
9
SN 1987A ACIS Spectrum (2-shock model)
NH 2.35 x 1021 cm-2 Soft component kTs
0.23 0.31 keV ne t 1013 cm-3 s Hard
component kTh 2.2 3.2 keV ne t 2 x
1011 cm-3 s ns/nh 20 Abundances fixed
at values obtained from the LETG data N 0.76
O 0.09 Ne 0. 29 Mg 0.24 Si 0.28 Si
0.45 Fe 0.16 (Zhekov et al. 2006)
2000 Jan 2005 July
2004-8 LETG
keV
10
SN 1987A Soft X-Ray Light Curve
Linear increase of X-ray flux until day 3000.
Rate jump in 1997 (day 3700) coincident with
emergence of optical spots. An exponential
radial density profile can fit the lightcurve
over a decade. An excess became evident since d
6200. Forward shock enters a wall?
0.5-2 keV 3-10 keV
d 3700
d 6200
X-ray (2005-7) vs. Optical (2005-4)
fx n02V0T0-0.6 nr2VrTr-0.6 Where V R3
R t2/3 (Chevalier 1982)
Tr T0 (n0/nr) (e-i equilibrium)
Filling factor of nr increases
e-(Rr-R)/D
0.5-2 keV fractional flux
0.4-1.5/2-5 keV ratio
Fast shock
fx ne2R3T-0.6(e.g., McKee Cowie 1977)
Where T const R3(n0nr e-(Rr-R)/s)2
X-ray Flux (10-13 ergs/cm2/s)
Image ACIS 0.5-2 keV Contours HST (Credit
Peter Challis)
d 6200
Slow shock
Chandra/ACIS
ROSAT (Hasinger et al. 1996)
Day since SN
11
SN 1987A X-Ray Light Curve Updates 2007-1
(Preliminary Results)
0.5-2 keV fractional flux
0.4-1.5/2-5 keV ratio
12
SN 1987A Hard X-Ray Light Curve
X-ray (2005-7) vs. Radio (2005-6)
Chandra (0.5 2 keV)
ATCA
X-ray Flux (10-13 ergs/cm2/s)
Image ACIS 3-8 keV Contours ATCA 9 GHz
ROSAT
Chandra (3 10 keV)
Similar rates of hard X-ray and radio The same
origin for them? Simply due to softening of
X-ray spectrum?
Image ACIS 0.4-0.5 keV Contours ATCA 9 GHz
Radio image B. Gaensler
L. Staveley-Smith
Day since SN
13
SN 1987A X-ray Radial Expansion
Racusin et al. 2007
X-ray radius vs time. The broadband
deconvolved image for each observation is
deprojected (43 deg) and fitted to 3 models (a
ring 4 spots) in order to estimate the radius
of the SNR as a function of time. Estimated
overall expansion velocity is 3900 km/s (with
poor fit). However, the expansion apparently
decelerates to 1400 km/s since d 6200.
14
SN 1987A Dispersed Spectrum (1999)
Combined Line Profile (day 4609)
Michael et al. 2002
Dispersed Spectrum (HETG) 1999
Vs 3400 700 km/s (consistent with radio
measurements) kTi 17 keV Observed kTe
2.5 keV
Line width 2300 300 km/s
(single Gaussian fit)

• Direct evidence for incomplete electron-ion
  thermal equilibration behind shock!

15
SN 1987A Dispersed Spectrum (2004)
Zhekov et al. 2005,2006
LETG/ACIS-S 289 ks in 2004-8/9 (day
6400) Detailed X-ray lines are resolved with
good photon statistics. kT 0.5 2.7
keV Doppler widths of individual lines are
measured v 300-1700 km/s LMC-like
abundances with N/O 1.1
Ne X
LETG 1st-order LETG 0th-order
Si XIII
Ne IX
O VIII
Mg XI
Fe XVII
Fe XVII
Mg XII
O VIII
Si XIV
N VII
O VII
V 340 1700 km/s (Where ?? 2??0 /-
2z0(?/?0?? ?)
Angstrom
FWHM (km/s)
OVIII
South (m 1)
North (m -1)
Angstrom
16
SN 1987A Blast Wave Arrives at Inner Ring!

• HST images optical spots dominate entire inner
  ring by
• 2003 (day 6000).
• Chandra images western side brightening up
  since 2003
• (day 6000) and now X-ray morphology is a
  complete ring.
• - Soft X-ray light curve makes a turn-up at day
  6200.
• Soft X-ray emission is now dominated by the
  decelerated
• shock since day 6000.
• - X-ray radial expansion rate reduces since day
  6200.
• - HETG/LETG Shock velocity reduces to 1400 km/s.
• Mid-IR intensity turns up since day 6000 dust
  emission
• from the inner ring (Bouchet et al. 2006).

mid-IR vs HST (2005-1)
17
SN 1987A Time-Lapse Movie
(2000-01 to 2005-01)
To be continued
Multi-wavelength collaboration
Chandra/HST/Spitzer/Gemini/ATCA Chandra AO8 740
ksec ACIS/LETG/HETG (PSU/Colorado/MIT)
observations accepted!
PSU/SAO/CXC
18
Comments on Harberl et al. (2006)

• ACIS pileup affects 20-25 of X-ray fluxes.
  Chandra results are thus artifacts.
• Generals
• - Corrections for days 4609, 6533 6716 are
  incorrect (HETG 1/8 subarray).
• - Basic assumptions are questionable.
• EPIC fluxes are correct or true ???
• Uncertainties on measurements for EPIC
  and ACIS are 1 ???
• (Harberl et al. determined their own
  ACIS fluxes and errors)
• Attribute discrepancy entire to the ACIS
  pileup ???
• EPIC vs. ACIS cross-calibrations
  several at 1-8 keV.
• ACIS CTI 5-10 CR
• ACIS QE Time-dependent, invalid
  constant CR-to-e flux conversion (up to 20).
• ACIS Pileup NOTE SN 1987A is an
  EXTENDED source with the ACIS!
• Specifics on ACIS pileup
• - Qualitative estimates significant effects on LC analysis.
• - Quantitative estimates
• 1. PIMMS/XSPEC simulations based on the
  XMM results 4711, 5176, 5980.
• 2. ACIS grade branching distribution
  analysis all epochs.
• 3. ACIS pileup modeling 4711, 5176,
  5980, 6157, 6533
• 
  (Not suitable for extended, line-dominated
  sources).

Conclusions The ACIS pileup does not affect our
scientific results. (Haberl et al. corrections
are incorrect and/or statistically/physically
insignificant.)