First XMM-Newton results on the Pulsar Wind Nebula in the composite supernova remnant G0.9+

a r X i v :a s t r o -p h /0204261v 1 16 A p r 2002

FIRST XMM-NEWTON RESULTS ON THE PULSAR WIND NEBULA IN THE COMPOSITE

SUPERNOV A REMNANT G0.9+0.1

D.Porquet 1,A.Decourchelle 1,and R.S.Warwick 2

1

Service d’Astrophysique,CEA Saclay,91191Gif-sur-Yvette Cedex,France

2

Department of Physics and Astronomy,University of Leicester,Leicester LE17RH,UK

Abstract

We report the results of the XMM-Newton observa-tion of the Pulsar Wind Nebula (PWN)in the composite supernova remnant G0.9+0.1located in the Galactic Cen-ter region.Thanks to the sensitivity of the EPIC cameras,we focus on the ?rst spectral analysis of large and small-scale structures of this PWN using MOS and PN data in combination.Our spatially resolved spectral analysis at large-scale o?ers a clear indication of a softening of the photon index with distance from the centroid of the neb-ula (from Γcore =1.12+0.45?0.48to Γ=2.42+0.19

?0.19),as observed in other known X-ray plerions.A spectral analysis of the small-scale structures in the central region of this PWN,delineates variations of the spectral index within the arc-like feature observed with Chandra :the eastern part has a hard photon index (Γ～1.2±0.5),while the south-west part has a softer photon index (Γ～2.8±0.7).

Key words:ISM:supernova remnants:individual:G0.9+0.1

–X–rays:ISM

medv9q20t5r615.rsp,m2

all ef20medium.rsp.The lo-cal background used corresponds to the central MOS 1

CCD region excluding the bright point sources and the central part within a radius of 2.6’.We subtract from the

source and the local background in our pointing,a blank-?eld observation (kindly provided by David Lumb)at the same position in order to take into account the particle background (see e.g.Majerowicz &Neumann 2001).The normalization between our pointing and the blank-?eld ?les is determined respectively by the count rate ratio in the 10–12keV range for the MOS and 12–14keV for the PN.The EPIC spectra were binned to give S/N ≥3.xspec v11.1.0was used for the spectral analysis.All sub-sequent errors are quoted at 90%con?dence.Abundances are those of Anders &Grevesse (1989).

3.The overall nebula

The XMM-Newton observation of the supernova remnant G0.9+0.1shows emission from the PWN,but not from the surrounding shell observed in radio.The soft X-ray emission expected from the shell is likely not seen due to the large interstellar absorption towards the source.The EPIC image of the PWN in G0.9+0.1is presented in Fig-ure 1.It exhibits a very bright central core,surrounded by extended di?use emission.The extent of this emission is about twice larger than observed with Chandra ,thanks to the high sensitivity of XMM-Newton .

For comparison with previous analysis,we ?t the spec-trum of the overall nebula.The region considered corre-sponds to the sum of the core with regions 1,2and 3,as

Figure1.XMM-Newton(EPIC)image in the3-8keV en-ergy band of the PWN in G0.9+0.1obtained with an adap-tative smoothing of signal to noise of5.This size of the image is2.7′×2.7′.The di?erent large-scale regions used for the spectral analysis are superposed.

is shown in Figure

1.The data are well?tted by either

a power-law or a thermal bremsstrahlung(Table1).For the photo-electric absorption,we use the cross-sections of Morrison&McCammon(1983).The parameters(N H,Γ)related to the power-law are compatible,within the er-ror bars,with the BeppoSAX values(Sidoli et al.2000:

N H=10.9+2.4

?2.1cm?2,Γ=1.95+0.33

?0.30

),though slightly higher

but more tightly constrained.In the following,we use the updated cross-sections for X-ray absorption by the ISM (tbabs in xspec)from Wilms et al.(2000).The param-eters although in good agreement with the previous ones (see Table1),give systematically slightly lower N H values using tbabs absorption model.

4.The large-scale structures

The XMM-Newton data enables us to study the variation of the spectral index within the PWN.In other plerions, a softening of the spectrum has been observed toward the outer regions like in3C58(Torii et al.2000,Bocchino et al.2001),G21.5-0.9(Slane et al.2000,Warwick et al. 2001)and IC443(Bocchino&Bykov2001).This soft-ening could be explained by the shorter lifetime of high energy electrons than for the lower energy electrons.

To search for this e?ect,we extracted the spectra in four regions,which are displayed in Figure1.The observed spectra and best-?t models for the regions1(top panel) and3(bottom panel)are shown,as examples,in Figure2. The spectrum extends up to almost12keV for the PN Table1.Results of the spectral?ts for the overall X-ray PWN(core and regions1,2,3in Fig.1).Cross-sections of the interstellar absorption are from Morrison&Mc-Cammon(1983;wabs in XSPEC)or from Wilms et al. (2000;tbabs).Uncertainties are quoted at90%con?-dence.The unabsorbed?uxes(2–10keV)are expressed in 10?12erg cm?2s?1.

Model N HΓor kTχ2νF unabs

2?10

(1023cm?2)(keV)(446dof)

tbabs*PL 1.39+0.13

?0.13

1.94+0.18

?0.18

375.2 5.69

tbabs*brems 1.27+0.10

?0.10

11.5+4.8

?2.6

375.0 5.12

tbabs*bb0.80+0.09

?0.08

1.75+0.09

?0.08

387.9 3.65

Figure2.Spectral?ts of the EPIC data(black for MOS1, red for MOS2and green for PN)with an absorbed power-law(tbabs*po).The inferred parameters are given in Ta-ble2.Top panel:Region1.Bottom panel:Region3.

data,and to about9keV in the MOS data allowing strong constraints on the determination of the photon index.

We?t the spectra of these four regions by absorbed power-laws,?xing N H to the value obtained for the overall PWN,i.e N H=1.431023cm?2(see Table1).The best-?t

24681012

distance from the centroid (unit of r 50)

Γ

Γ?radius relation

Figure 3.Variation of the spectral index versus distance to the centroid of the nebula.The X-axis shows the weighted mean distance of the pixels in a given region from the cen-troid expressed in unit of r 50,the radius at which the ple-rion surface brightness drops by a factor of 2(7.15′′for G0.9+0.1),as de?ned in Bocchino &Bykov (2001).For comparison,data for IC 443are also plotted (taken from Bocchino &Bykov 2001).

parameters are given in Table 2.Letting the absorption column densities free do not a?ord signi?cantly better ?ts,and the values of column density N H are compatible to within 15%for the di?erent regions of the nebula.

A clear steepening of the spectrum is observed from the inner part toward the outer part of the PWN as is shown in Figure 3.The observed spectral softening with radius in G0.9+0.1from the core to the outskirts of the PWN,is

Table https://www.sodocs.net/doc/521642878.html,bined ?ts of the regions of the PWN in G0.9+0.1.The cross-sections of the interstellar absorp-tion are from Wilms et al.(2000).Unabsorbed X-ray ?ux (2-10keV)are expressed in 10?12ergs cm ?2s ?1.Top:N H is frozen to the value obtained for the overall PWN (see Table 1).Bottom:N H is a free parameter.

N (a )

H

Γ

χ2ν/dof F (b )

X

(1023cm ?2)(2-10keV)

core 1.39 1.12+0.45

?0.48

15.8/18

0.25region 1 1.39 1.46+0.14

?0.14

103.4/146 1.79region 2 1.39 2.02+0.17

?0.17

121.6/136 1.65region 3 1.39

2.42+0.19

?0.19

161.8/173 2.02core 1.45+0.88

?0.66

1.18+1.17

?0.99

15.8/170.26region 1 1.53+0.25

?0.22

1.62+0.30

?0.28

102.3/145 1.93region 2 1.54+0.27

?0.23

2.21+0.38

?0.34

120.6/135 1.83region

3

1.35+0.24

?0.21

2.36+0.39

?0.36

161.7/172

1.95

most likely due to the e?ect of synchrotron radiation losses electrons as they progress through comparison we have reported on Figure 3the spectral index inferred from the ob-443by Bocchino &Bykov (2001).The 50,as de?ned in Fig.3and in Bocchino gives a measure independent on the dis-Our data appear to show a stronger core to the outer part of the nebula to IC 443(see Fig.3),and also to such as 3C 58and G21.5-0.9(see for com-Bocchino &Bykov 2001).This could be a stronger magnetic ?eld in the Galactic to other SNR environments.

small-scale structures

The brightest region of the nebula corresponds to a core of angular diameter 8′′as observed with XMM-Newton .The Chandra observation has resolved it as an elliptical clump of size 5′′×8′′(Gaensler et al.2001).The faint point source proposed to be the pulsar by Gaensler et al.(2001)is outside this region (about 8′′from the center

[0.6-1.6]

1.1

[0.8-1.7]

1.2

[1.2-2.4]

1.8

[1.0-2.4]

1.7

[2.1-3.4]

2.8

[1.2-2.2]

1.7

[1.2-2.1]

1.6

[1.3-2.2]

1.7

Figure 4.Close-up view with XMM-Newton (EPIC)of the center of the nebula in the 3-8keV energy band obtained with an adaptative smoothing of signal to noise of 5.The size of the image is 1.7′×1.3′.The contours (in black)corresponds to the Chandra observation in the same en-ergy range.The white boxes represent di?erent small-scale structures surrounding the bright core (dark blue circle),for which the spectral index and 90%con?dence uncertain-ties are given.The position of the structures were deter-mined according to the Chandra observation (see Gaensler et al.2001):“East arc”(Γ～1.2),“jet”(Γ～1.8),and “South-West arc”(Γ～2.8).

of the clump).The fact that the hardest spectral index is obtained for the core would likely indicate that it hosts the pulsar.As proposed by Gaensler et al.(2001),this clump may also be an analog to the knots observed in the Crab nebula(Hester et al.1995)and close to the Vela pulsar(Pavlov et al.2001),which exhibit variability in brightness,position and morphology.The position and the morphology are however very similar in both XMM-Newton and Chandra observations which are separated from only about one month.The nature of the bright core is still unclear:a clump,which could be a region of sep-arate particle acceleration as in PSR B1509-58(Gaensler et al.2002),or a very compact region in which the pulsar is embedded?.

Tracing the spectral variations in the small-scale struc-tures around the core is then essential for understanding this particular PWN.Figure4displays a closer view of the inner region and shows the small-scale structures present around the core of PWN G0.9+0.1.For comparison,the X-ray contours of the Chandra data are superposed.In our data,the small-scale structures are not so clearly detected as with Chandra(jet-like and arc-like features).However using the position of the structures seen by Chandra and the high sensitivity of XMM-Newton,we were for the?rst time able to carry out spectral analysis of these small-scale structure.Determining their spectral index gives an indication of the geometry and orientation of the nebula. Figure4reports the photon spectral index of the corre-sponding structures obtained with combined?ts(MOS and PN).The region corresponding to the eastern part of the arc-like feature in Gaensler et al.(2001),shows the hardest spectral index(Γ～1.2±0.5)among all the struc-tures together with the core region(Γ～1.1±0.5).On the contrary the region corresponding to the south-west part of the arc-like feature,appears to have the steepest photon index(Γ～2.8±0.7)over all the PWN.A possible explana-tion is that the east arc is pointing towards the observer and that its spectral hardness is due to the relativistic beaming or Doppler boosting of the electrons,while the south-west arc is the opposite part of the torus.The region associated to the jet-like feature in Gaensler et al.(2001) does not exhibit a harder spectrum as suggested by their hardness ratio.

6.Conclusion

We report XMM-Newton observations of the X-ray Pul-sar Wind Nebula(PWN)inside the composite supernova remnant G0.9+0.1(Helfand&Becker1987),which was convincingly detected for the?rst time in X-rays by Bep-poSAX(Mereghetti et al.1998).

The large-scale image of the PWN G0.9+0.1shows a very bright central core,surrounded by relatively large-scale di?use emission with a much spatial extent than is apparent for the recent Chandra images.Thanks to its high sensitivity,XMM-Newton allows for the?rst time a spectral analysis of the large and small-scale structures (see also Porquet,Decourchelle&Warwick2002).Our spectral analysis combining the MOS and PN data o?ers

a clear indication of a softening of the spectrum with dis-tance from the centroid(related to the?nite lifetime of the synchrotron emitting electrons),as observed in other known X-ray plerions.Even if in our data,the small-scale structures are not so clearly detected as in the Chandra data(jet-like and arc-like features),we were able for the ?rst time to carry out their spectral analysis.Two of the structures show distinct spectral indexes from the gen-eral trend(fromΓ～1.6±0.5toΓ～1.8±0.6):the eastern part of the arc-like feature has a very hard photon index (Γ～1.2±0.5),while the south-west part of the arc-like feature has a softer photon index(Γ～2.8±0.7).

The study of G0.9+0.1provides further clues concern-ing the processes by which pulsars connect with their en-vironment and also illustrates the impact of a strong am-bient magnetic?eld as in the Galactic Center region.

Acknowledgements

D.P.acknowledges grant support from the“Institut National des Sciences de l’Univers”and from the“Centre National d’Etudes Spatial”(France).

References

Anders E.&Grevesse N.,1989,Geochimica et Cosmochimica Acta,53,197

Bocchino F.&Bykov A.M.,2001,A&A,376,248

Bocchino F.,Warwick,R.S.,Marty,P.,Lumb,D.,Becker,W., Pigot,C.2001,A&A,369,1078

Gaensler,B.M.,Pivovaro?M.J.,Garmire G.P.2001,ApJ,556, L107

Gaensler, B.M.,Arons,J.,Kaspi,V.M.,Pivovaro?,M.J., Kawai,N.,Tamura,K.2002,ApJ,569,in press

Helfand D.J.&Becker R.H.,1987,ApJ314,203

Hester J.J.,Scowen P.A.,Sankrit R.,et al.,1995,ApJ,448, 240

Majerowicz,S.&Neumann D.M.2001,in the“XXI Moriond Conference”,Eds. D.Neumann, F.Dur-ret,&J.Tran Thanh Van,in press(http://www-

dapnia.cea.fr/Conferences/Morion2001/index.html) Mereghetti S.,Sidoli L.,Israel G.L.1998,A&A,331,L77 Morrison R.&McCammon D.,1983,ApJ,270,119

Pavlov,G.G.,Kargaltsev,O.Y.,Sanwal,D.,Garmire,G.P.

2001,ApJ,554,L189

Porquet,D.,Decourchelle A.,Warwick R.S.,2002,in prepara-tion

Sidoli L.,Mereghetti S.,Israel G.L.,Bocchino F.,2000,A&A, 361,719

Slane P.,Chen,Y.,Schulz,N.S.,Seward,F.D.,Hughes,J.P., Gaensler,B.M.2000,ApJ,533,L29

Torii K.,Slane,P.O.,Kinugasa,K.,Hashimotodani,K., Tsunemi,H.2000,PASJ,52,875

Warwick R.S.,Bernard,J.-P.,Bocchino,F.,Decourchelle,A., Ferrando,P.,Gri?ths,R.G.,Haberl,F.,La Palombara,

N.,Lumb,D.,Mereghetti,S.,Read,A.M.,Schaudel,D., Schurch,N.,Tiengo,A.,Willingale,R.2001,A&A,365,

L248

Wilms J.,Allen A.,McCray R.,2000,ApJ,542,914