Table Of ContentMon.Not.R.Astron.Soc.000,1–??(2011) Printed17January2012 (MNLATEXstylefilev2.2)
Suzaku study of centrally brightened supernova remnant
G272.2−3.2
A. Sezer,1,2(cid:63) F. G¨ok 3 (cid:63)†
2 1TU¨BI˙TAK Space Technologies Research Institute, ODTU Campus, Ankara, 06531, Turkey
1 2Bo˜gazic.i University, Faculty of Art and Sciences, Department of Physics, I˙stanbul, 34342, Turkey
0 3Akdeniz University, Faculty of Sciences, Department of Physics, Antalya, 07058, Turkey
2
n
a
J
6
1 ABSTRACT
In this work, the results from Suzaku observation of Galactic supernova rem-
] nant G272.2−3.2 are presented. Spectra of G272.2−3.2 are well fitted by a single-
E
temperature variable abundances non-equilibrium ionization (VNEI) model with an
H electron temperature kT ∼0.77 keV, ionization timescale τ ∼6.5×1010 cm−3s and
e
. absorbing column density N ∼ 1.1×1022 cm−2. We have detected enhanced abun-
h H
dances of Si, S, Ca, Fe and Ni in the center region indicating that the X-ray emission
p
has ejecta origin. We estimated the electron density n to be ∼0.48f−1/2 cm−3, age
- e
o ∼4300f1/2 yr and the X-ray total mass M =475f1/2 M(cid:12) by taking the distance to
x
r be d=10 kpc. To understand the origin of the centrally-peaked X-ray emission of the
t
s remnant, we studied radial variations of the electron temperature and surface bright-
a
ness. The relative abundances in the center region suggest that G272.2−3.2 is the
[
result of a Type Ia supernova explosion.
1
Key words: ISM: supernova remnants−ISM:individual G272.2−3.2−X-rays:ISM
v
4
2
2
3
1 INTRODUCTION correlate well with the brightest optical filaments and of a
.
1 diffuse emission component produced by shock-accelerated
0 The Galactic supernova remnant (SNR) G272.2−3.2 was electrons. Harrus et al. (2001) found that the X-ray emis-
2 discovered in X-rays with ROSAT All-Sky Survey (Greiner
sion from G272.2−3.2 could be best described by a non-
1 &Egger1993),ithascentrallyfilledX-raymorphologyand
equilibrium ionization (NEI) model with an electron tem-
: a thermally dominated X-ray spectrum. Greiner, Egger &
v perature of about 0.7 keV, an ionization timescale of 3200
Aschenbach(1994)foundthattheelectrontemperaturewas
i cm−3 yr and a relatively high column density, N ∼ 1022
X between 1.0−1.5 keV from the ROSAT PSPC data. Win- H
atoms−2, from ASCA and ROSAT observations. They dis-
r kler,Hanson&Phillips(1993)confirmedthatthenatureof cussed cloud evaporation and thermal conduction models
a the emission was shock-heated and the nebulosity was an
to explain the centrally peaked X-ray morphology of the
SNR by measuring the [Sii]/Hα ratio and detecting emis-
remnant. Later, from an X-ray study of G272.2−3.2 with
sions from [Nii] 658.3 nm and [Oii] 732.5 nm in the opti-
Chandra,Parketal.(2009)reportedthattheX-rayspectra
cal band. Duncan et al. (1997) carried out a series of ra-
of the outer shell regions showed normal compositions be-
dioobservationswithaParkesradiotelescope,anAustralia
ing consistent with the shocked interstellar medium, while
Telescope Compact Array (ATCA) and a Molonglo Obser-
centralemissionshowedelevatedabundancessuggestingre-
vatory Synthesis Telescope (MOST). From these observa-
verse shocked stellar ejecta from Type Ia supernova (SN).
tions they show that the radio spectral index of this rem-
Lopez et al. (2011), based on Chandra’s observation, con-
nant was typical of shell-type SNRs, α ∼ 0.55, and almost
firmed that G272.2−3.2 is a Type Ia origin.
circular in appearance with a diameter of ∼15 arcmin. The
remnant consists of faint filaments and patches of emission The distance to G272.2−3.2 is not well known. It was
with a low surface brightness as well as bright blobs that estimated to be 1.8+−10..48 kpc from observed interstellar ab-
sorption by Greiner, Egger & Aschenbach (1994). From the
statistical analysis, Harrus et al. (2001) calculated 2 kpc,
which is in agreement with the distance found by Greiner,
(cid:63) E-mail: [email protected] (AS);
Egger & Aschenbach (1994). They also obtained an upper
[email protected](FG)
† This file has been amended to highlight the proper use of limit of 10 kpc using optical color excess with a distance of
LATEX2εcodewiththeclassfile.Thesechangesareforillustrative roughly 0.2 mag kpc−1 in the direction of G272.2−3.2 and
purposesanddonotreflecttheoriginalpaperbyA.Sezer. adopted intermediate distance of 5 kpc to this remnant.
(cid:13)c 2011RAS
2 A. Sezer, F. G¨ok
In this study, we investigate the nature of X-ray emis- thisisindicatedbysolidblackcircles(seeFig.1)foreachof
sion of G272.2−3.2 which is characterized by an apparent the XISs. The spectra are grouped with a minimum of 120
centrallybrightenedX-raymorphologyandthermallydom- counts bin−1 for the whole region and 50 counts bin−1 for
inated X-ray emission by utilizing the superior spectral ca- the central and outer regions.
pabilitiesfordiffusesourcesofX-rayImagingSpectrometers For the whole region, we applied the VNEI model, a
(XIS: Koyama et al. (2007)) onboard Suzaku satellite (Mit- model in xspec for a NEI collisional plasma with variable
suda et al. 2007). The structure of the paper is as follows; abundances(Borkowski,Lyerly&Reynolds2001),modified
in Section 2, we describe the Suzaku observation and data byinterstellarabsorptionusingcrosssectionsfromMorrison
reduction.ImageandspectralanalysisarepresentedinSec- &McCammon(1983)inthe0.3−10keVenergyrange.The
tion3and4,respectively.InSection5,wediscussthephys- absorption column density N , electron temperature kT ,
H e
ical properties of the thermal X-ray emitting plasma (5.1), ionization timescale τ = n t, where n is the electron den-
e e
possible reasons of centrally peaked morphology (5.2) and sityandtistheelapsedtimeaftertheplasmawasheatedup,
finally, relative abundances in the ejecta (5.3). and normalization were set as free parameters and all ele-
mentalabundanceswerefixedtotheirsolarvaluesofAnders
& Grevesse (1989). From this fit we obtained a reduced χ2
2 OBSERVATION AND DATA REDUCTION of 1.81 for 752 degrees of freedom (dof). Then, we allowed
O, Ne, Mg, Si, S and Fe abundances to vary, since these
Suzaku observed G272.2−3.2 on 2011 May 28 by the XIS.
abundances have appeared to differ from their solar values
The observation ID and exposure time are 506060010 and
andthelinefeatureswereevidentinthespectra,whileother
130 ksec, respectively. The XIS has four CCDs: three of
elemental abundances were frozen to their solar values. In
them (XIS0, 2, and 3) are front-illuminated (FI) and one
thiscase,thespectralfithassignificantlyimprovedwithre-
(XIS1) is back-illuminated (BI). The XISs are sensitive to duced χ2 of 1.02 for 746 dof. We repeated the same steps
the 0.2−12.0 keV energy band with a 17.8×17.8 arcmin2
for the central and outer regions. The parameter values ob-
field of view (FOV). In November of 2006, XIS2 was dam-
tained for each region are listed in Table 1, and the errors
aged and taken off-line, therefore data taken after the 2007
quoted are 90 per cent confidence limits. The background-
observationweretakenwithonlytheremainingthreeXISs.
subtracted FI XIS (XIS0 and XIS3) spectra of each region
The XIS was operated in the normal full-frame clocking
in 0.3−10 keV are shown in Fig. 2.
mode. Two corners of each XIS CCD have an 55Fe calibra-
Radial variations of the electron temperature and sur-
tionsourcewhichcanbeusedtocalibratethegainandtest
facebrightnessareplottedinFig.3.Duringspectralfittings,
the spectral resolution of data taken using this instrument.
to get an estimate of possible temperature variation across
Reduction and analysis of the data were performed by
the SNR, we fixed the absorbing column density and the
following the standard procedure using the headas v6.4
ionization timescale to the values of the whole region.
software package, and spectral fitting was performed with
xspec v.11.3.2 (Arnaud 1996). All of the data were repro-
cessed, referring to the CALDB as of July 9, 2008. The re-
5 DISCUSSION AND CONCLUSIONS
distribution matrix files (RMFs) of the XIS were produced
by xisrmfgen, and auxillary response files (ARFs) by xis- In this paper, we report the results of high quality X-ray
simarfgen (Ishisaki et al. 2007). spectra and detailed analysis of G272.2−3.2 using Suzaku
XISobservation.TheX-rayspectracanberepresentedwith
anon-equilibriumionizationplasma(VNEI)modelwithan
3 IMAGE ANALYSIS electrontemperatureofkT ∼0.77keV,highabsorbingcol-
e
umndensityandarelativelysmallionizationtimescale,less
Fig. 1 shows an XIS1 image of G272.2−3.2 in the 0.3−10
than1012 cm−3s.WefoundclearK-shelllinesofO,Ne,Mg,
keV energy band. From this figure, we see brighter emis-
Si, S, Ca, Fe and Ni in the 0.3−10 keV band spectra, as
sion in the central region (within ∼3.8 arcmin radius) and
shown in Fig. 2. The central region is enhanced in Si, S,
a relatively fainter emission in the outer part. Central and
Ca,FeandNi,asshowninTable1.Thisfactsuggeststhat
outer regions are shown by solid black circles centered at
theX-rayemissionoriginatingfromthisregionresultsfrom
RA(2000)=09h06m47s,Dec.(2000)=−52◦06(cid:48)05(cid:48)(cid:48).Dashed
the ejecta. The abundances of the outer region are consis-
whitecircleswithsizesof0−1.5,1.5−2.5,2.5−3.5,3.5−4.5,
tent with solar values indicating that the X-ray emission is
4.5−5.5 arcmin are chosen to obtain the radial variations
producedbyswept-upinterstellarmatter(ISM).Theabun-
of the electron temperature kT and surface brightness.
e dances obtained from the spectra of the whole region show
The black dashed circle with a radius of 1.5 arcmin rep-
thattheX-rayemissionresultsfromamixtureoftheejecta
resents the background region, RA(2000) = 09h07m18s,
andISM.ThisremnantshowsacentrallypeakedX-rayemis-
Dec. (2000) = −52◦14(cid:48)27(cid:48)(cid:48), used for spectral analysis. The
sion and extends to a radius of ∼7.3 arcmin as seen in the
black dashed square indicates the FOV of the XIS1.
XIS1image(seeFig.1).Wewilldiscusthepossiblereasons
for the centrally peaked emission in subsection 5.2.
4 SPECTRAL ANALYSIS
5.1 Thermal emission
The spectrum is extracted first from all over the remnant
(hereafter whole region) with a radius of 7.3 arcmin, then From the thermal X-ray spectra of G272.2−3.2, we found
from the central region with the brightest X-rays and fi- a high absorbing column density of N ∼1.07×1022cm−2
H
nally from the outer region where the X-rays are fainter, that is in agrement with the value obtained by Harrus et
(cid:13)c 2011RAS,MNRAS000,1–??
Suzaku observation of G272.2−3.2 3
Table 1. Best-fitting parameters of the spectral fitting in the 0.3−10 keV energy band for all over the
remnant (whole), its centre and outer regions with an absorbed VNEI model with corresponding errors at
90percentconfidencelevel(2.7σ).
Parameters Whole Centre Outer
NH(×1022cm−2) 1.07±0.02 1.14±0.02 0.96±0.02
kTe(keV) 0.77±0.02 0.83±0.03 0.76±0.02
O(solar) 1.4±0.4 (1) 0.6±0.2
Ne(solar) 0.6±0.1 0.2±0.1 0.4±0.1
Mg(solar) 0.7±0.1 0.7±0.1 0.6±0.1
Si(solar) 1.3±0.1 2.0±0.1 0.8±0.1
S(solar) 2.2±0.2 4.0±0.2 1.2±0.1
Ca(solar) (1) 1.8±1.1 (1)
Fe(solar) 1.3±0.2 1.96±0.11 0.8±0.1
Ni(solar) (1) 3.9±0.9 (1)
net(×1010cm−3s) 6.5±0.6 5.3±0.5 6.2±0.6
normalization 0.19±0.02 0.17±0.01 0.16±0.01
Fluxa 6.8±0.1 7.7±0.1 4.3±0.1
χ2/dof 760.7/746=1.02 917.2/686=1.34 733.7/945=0.78
a FluxcorrectedforGalacticabsorptioninthe0.3−10keVenergybandintheunitof10−11ergs−1cm−2.
al. (2001) with ROSAT spectral fits and with the Galactic 5.2 Radial profile
Hi column density in that direction, N ∼0.9×1022cm−2
H Centrally peaked X-ray morphology of G272.2−3.2 can be
(Dickey & Lockman 1990). During the analysis we let N
H explained by two models cloud evaporation (White & Long
vary for whole, center and outer parts to see if there is a
1991) and thermal conduction (Cox et al. 1999). We first
significant variation all over the remnant. We found that
consider the cloud evaporation model of White & Long
the N value obtained for these three regions are similar,
H (1991).Accordingtothismodel,theSNRblastwavepasses
indicating that there is not a significant density gradient
overthecoldcloudskeepingtheminthehotpostshockgas.
acrosstheremnant.OnereasonforthehighN valuemight
H X-ray emission arises from the gas evaporated from these
be that the remnant’s distance is large or there could be
shocked clouds. Second, we consider the radiative model
molecular material or dust along the line of sight in that
of Cox et al. (1999) also named as the “fossil” conduction
direction, or both. However, in literature no such material
model. According to this model the hot plasma in the inte-
has been reported yet in that direction or in the vicinity of
riorsoftheremnantgraduallybecomesuniformbythermal
theremnant.Consideringthesecases,wewillusetheupper
conduction and detectable as centrally brightened in X-ray.
limitdistancevalue,d=10kpc,givenbyHarrusetal.(2001)
We studied radial variations of the electron tempera-
throughout our calculations.
ture and surface brightness profiles (see Fig. 3) to com-
pare with these two models. We see that there is no
strong radial temperature variation (∼0.06 keV) and it
is consistent with the predictions of both evaporation
and thermal conduction models. Observed surface bright-
The XIS spectra suggest an ionization time scale of
ness variation which peaks at the center (∼1.26×10−11
n t ∼ 6.5 × 1010 cm−3s. For the full ionization equilib-
e erg s−1cm−2arcmin−2) and declines towards outer region
rium, the ionization timescale, τ=n t, is required to be
e (∼0.38×10−11 ergs−1cm−2arcmin−2)isconsistentwiththe
(cid:62)1012 cm−3s (Masai 1984). The value that we obtained for
evaporationmodel.However,inthevicinityofthisremnant
G272.2−3.2showsthattheplasmaisfarfromthefullioniza-
nomolecularcloudsordensitygradientofmediumhasbeen
tion equilibrium. We estimated the X-ray emitting plasma
reportedyet.Thiscaseisinconsistentwiththepredictionsof
volume of the remnant to be ∼1.2×1060f cm3, where f
bothmodels.Theyoungage(∼4300f1/2yr)oftheremnant,
is the volume filling factor of the emitting gas within the
in other words the NEI condition of the plasma can not be
SNR, we assumed the emitting region to be a full sphere of
explained by the thermal conduction model which requires
radius 7.3 arcmin which is our XIS spectral extraction re-
collisional ionization equilibrium condition of the plasma.
gion. Based on the emission measure EM = n n V deter-
e H Therefore, considering all these, the centrally peaked X-ray
mined by the spectral fitting, where n and n are number
e H emission of this remnant may be explained with the cloud
densities ofelectrons and protons respectively, and Vis the
evaporation model.
X-rayemittingvolume,andassumingn =1.2n ,wecalcu-
e H
latedtheelectrondensityoftheplasman tobe∼0.48f−1/2
e
cm−3. The age of G272.2−3.2 calculated to be ∼4300f1/2
5.3 Relative abundances in the ejecta
yr from t=τ/n . The mass of the X-ray emitting plasma of
e
G272.2−3.2 estimated to be M =475f1/2 M(cid:12) from equa- TypeIaSNproducesverysmallquantitiesoflow-Zelements
x
tion M = n Vm , where m is the mass of a hydrogen suchasNeandMg,andlargeramountsofSi-groupelements
x e H H
atom. suchasSandCa,andoverabundantFeandNiasinourcase
(cid:13)c 2011RAS,MNRAS000,1–??
4 A. Sezer, F. G¨ok
withG272.2−3.2.Therefore,wecompareourbest-fittingrel- Yamaguchi H. et al., 2008, PASJ, 60, 141
ativeabundanceswiththepredictednucleosynthesisyieldof
thewidely-usedW7andadelayeddetonation(WDD2)Type
Ia SN models (Nomoto et al. 1997) as given in Fig. 4.
AlthoughtheabundancesofNe,MgandCarelativeto
Siarealmostconsistentwithbothmodels,SandNirelative
to Si are higher than the values that both models predict.
The value of Fe relative to Si is consistent with the WDD2
model, while it is much lower than the value of W7 model
predicts.ThereasonforthelowvalueofFemightbethatthe
entire Fe-rich core has not yet been shocked, as is the case
in SN 1006 (Yamaguchi et al. 2008), Tycho (Tamagawa et
al.2009)andG337.2-0.7(Rakowski,Hughes&Slane2001).
Our results confirm that G272.2−3.2 has a Type Ia SN ori-
gin.
ACKNOWLEDGMENTS
ASissupportedbytheTU¨BI˙TAKPostDoctoralFellowship.
ThisworkissupportedbytheAkdenizUniversityScientific
Research Project Management.
REFERENCES
Anders E., Grevesse N., 1989, Geochimica Cosmochimica
Acta, 53, 197
Arnaud K. A., 1996, in Jacoby G., Barnes J., eds, ASP
Conf. Ser. Vol.101, Astronomical Data Analysis Software
and Systems V. Astron.Soc. Pac., San Francisco, p. 17
Borkowski K. J., Lyerly W. J., Reynolds S. P., 2001, ApJ,
548, 820
Cox D. P., Shelton R. L., Maciejewski W., Smith R. K.,
Plewa T., Pawl A., R´oy˙yczka M., 1999, ApJ, 524, 179
Dickey J.M., Lockman F. J., 1990, ARA&A, 28, 215
DuncanA.R.,PrimasF.,RebullL.M.,BoesgaardA.M.,
Deliyannis C. P., Hobbs L. M., King J. R., Ryan S. G.,
1997, MNRAS, 289, 97
Greiner J., Egger R., 1993, IAU Circ. 5709
GreinerJ.,EggerR.,AschenbachB.,1994,A&A,286,L35
HarrusI.M.,SlaneP.O.,SmithR.K.,HughesJ.P.,2001,
ApJ, 552, 614
Ishisaki Y. et al., 2007, PASJ, 59, 113
Koyama K. et al., 2007, PASJ, 59, 23
LopezL.A.,Ramirez-RuizE.,HuppenkothenD.,Badenes
C., Pooley D.A., 2011, ApJ, 732, 114
Masai K., 1984, Ap&SS, 98, 367
Mitsuda K. et al., 2007, PASJ, 59, 1
Morrison R., McCammon D., 1983, ApJ, 270, 119
Nomoto K., Iwamoto K., Nakasato N., Thielemann F.-K.,
BrachwitzF.,TsujimotoT.,KuboY.,KishimotoN.,1997,
Nuclear Physics A, 621, 467
Park S., Lee J., Hughes J.P., Slane P.O., Burrows D.N.,
MoriK.,GarmireG.P.,2009,AmericanAstronomicalSo-
ciety, 41, 695
Rakowski C. E., Hughes J. P., Slane P., 2001, ApJ, 548,
258
Tamagawa T. et al., 2009, PASJ, 61, 167
White R. L., Long K. S., 1991, ApJ, 373, 543
Winkler P. F., Hanson G. J., Phillips M.M., 1993, IAU
Circ. 5715
(cid:13)c 2011RAS,MNRAS000,1–??
Suzaku observation of G272.2−3.2 5
Figure 1.SuzakuXIS1imageofG272.2−3.2inthe0.3−10keVenergyband.Solidblackcirclesshowcentralandouterregions,dashed
whitecirclesshowtheregionschosentoobtaintheradialvariationsoftheelectrontemperatureandsurfacebrightness.Theblackdashed
circlerepresentsthebackgroundregion.TheFOVoftheXIS1isindicatedbytheblackdashedsquare.Thecoordinates(RAandDec.)
arereferredtoepochJ2000.
(cid:13)c 2011RAS,MNRAS000,1–??
6 A. Sezer, F. G¨ok
Figure 2. Suzaku FI (XIS0:red and XIS3:black) spectra of G272.2−3.2 in the 0.3−10 keV energy band. (a) Whole region, (b) Central
region,(c)Outerregion.Thebottomwindowsgivetheresidualsfromthebest-fittingmodelforFIXISspectra.
Figure 3.RadialvariationsofobservedelectrontemperatureandsurfacebrightnessofG272.2−3.2.Surfacebrightnessisintheunitof
(×10−11)ergs−1cm−2arcmin−2.
(cid:13)c 2011RAS,MNRAS000,1–??
Suzaku observation of G272.2−3.2 7
Figure 4. Best-fitting abundance ratios of Ne, Mg, Si, S, Ca, Fe and Ni relative to Si are shown by crosses and predicted abundance
ratiosfromthecarbondeflagration(W7,Nomotoetal.(1997))modelareshownbydiamondsandfromthedelayeddetonation(WDD2,
Nomotoetal.(1997))modelareshownbysquares.
(cid:13)c 2011RAS,MNRAS000,1–??
