A detailed study of non-thermal X-ray properties and interstellar gas toward the \gamma-ray supernova remnant RX J1713.7-3946

1.3 MB·

by  H. Sano

accepted version (9th November2014)
PreprinttypesetusingLATEXstyleemulateapjv.08/22/09
A DETAILED STUDY OF NON-THERMAL X-RAY PROPERTIESAND INTERSTELLAR GAS TOWARD THE
γ-RAY SUPERNOVA REMNANT RX J1713.7 3946
−
H. Sano1, T. Fukuda1, S. Yoshiike1, J. Sato1, H. Horachi1, T. Kuwahara1, K. Torii1, T. Hayakawa1, T. Tanaka2,
H. Matsumoto1, T. Inoue3, R. Yamazaki3, S. Inutsuka1, A. Kawamura4, H. Yamamoto1, T. Okuda4, K.
Tachihara1, N. Mizuno4, T. Onishi5, A. Mizuno6, F. Acero7, and Y. Fukui1
1DepartmentofPhysics,NagoyaUniversity,Furo-cho,Chikusa-ku,Nagoya464-8601, Japan;[email protected]
4 2DepartmentofPhysics,KyotoUniversity,Kitashirakawa-oiwake-cho, Sakyo-ku,Kyoto606-8502,Japan
3DepartmentofPhysicsandMathematics, AoyamaGakuinUniversity,Fuchinobe,Chuou-ku,Sagamihara252-5258,Japan
1 4NationalAstronomicalObservatoryofJapan,Mitaka181-8588, Japan
0 5DepartmentofAstrophysics,GraduateSchoolofScience,OsakaPrefectureUniversity,1-1Gakuen-cho,Naka-ku,Sakai599-8531,Japan
2 6Solar-TerrestrialEnvironmentLaboratory, NagoyaUniversity,Furo-cho,Chikusa-ku,Nagoya464-8601,Japanand
7LUPM,CNRS/Universit´eMontpellier2,France
v
accepted version (9th November 2014)
o
N
ABSTRACT
0 We have carried out a spectral analysis of the Suzaku X-ray data in the 0.4–12 keV range toward
1 theshell-typevery-high-energyγ-raysupernovaremnantRXJ1713.7 3946. Theaimsofthisanalysis
−
are to estimate detailed X-rays spectral properties at a high angular resolution up to 2 arcmin, and

] to compare them with the interstellar gas. The X-ray spectrum is non-thermal and used to calculate
E absorbing column density, photon index, and absorption-corrected X-ray flux. The photon index
H varies significantly from 2.1 to 2.9. It is shown that the X-ray intensity is well correlated with the
. photon index, especially in the west region,with a correlationcoefficient of 0.81. The X-ray intensity
h tends to increase with the averaged interstellar gas density while the dispersion is relatively large.
p
The hardest spectra having the photon index less than 2.4 are found outside of the central10 arcmin
o- ofthe SNR,fromthe northtothesoutheast( 430arcmin2)andfromthe southwesttothe northwest
∼
r ( 150arcmin2). Theformerregionshowslowinterstellargasdensity,whilethelatterhighinterstellar
st g∼as density. We presentdiscussionfor possible scenarioswhich explain the distribution of the photon
a index and its relationship with the interstellar gas.
[ Subject headings: cosmic rays – ISM: clouds – ISM: individual objects (RX J1713.7 3946) – ISM:

supernova remnants – X-rays: ISM −
4
v
8
1. INTRODUCTION (2010)andGabici & Aharonian(2014). Eventhoughthe
1
Leptonic scenariosare still debated from a different per-
4 Supernova remnants (SNRs) in the Galaxy are be-
spective (e.g., Ellison et al. 2012).
7 lieved to be most likely accelerators of cosmic-rays
. (CRs) below “Knee’’, in an energy range less than The young SNR RX J1713.7 3946 (also known
1 as G347.3 0.5) is one of the be−st targets for test-
0 3 1015 eV. The highest energy γ-rays in the Galaxy −
× ing the action of the CR electrons, because the
4 are emitted from three young and very-high-energy
SNR emits strong synchrotron X-rays and VHE γ-
1 (VHE; E > 100 GeV) γ-ray SNRs, RX J1713.7 3946
− rays (Pfeffermann & Aschenbach 1996; Enomoto et al.
: (Aharonian et al. 2004, 2006b, 2007), RX J0852.0 4622
v − 2002; Aharonian et al. 2004, 2006a,b, 2007). RX
(Aharonian et al. 2005, 2007), and HESS J1731 347
i − J1713.7 3946 has a large apparent diameter of 1 de-
X (H.E.S.S. Collaboration et al.2011). IntheseSNRs,itis − ∼
gree at a small distance of 1 kpc (Fukui et al. 2003).
r suggested that CR protons or electrons are accelerated Subsequently, the X-ray prop∼erties of RX J1713.7 3946
a efficiently up to 10–800 TeV or 1–40 TeV, respectively −
were studied by using XMM-Newton, Suzaku, and
(e.g.,Zirakashvili & Aharonian2010). Allthethreeshow
Chandra. Chandra observations revealed filamentary
pure synchrotron X-rays with no thermal features, and
structures of X-rays and short time variability ( 1 yr)
accelerationofTeVCRelectronsiswellestablished. The ∼
of one of them (Uchiyama et al. 2003, 2007), which is
originoftheγ-raysintheseSNRs,eitherhadronicorlep-
interpreted as due to rapid particle acceleration of CR
tonic,hasbeenunderdebate;theGeVγ-rayobservations
electrons and cooling in 1 yr scale with mG magnetic
showhardγ-rayspectrumsimilarto that ofthe leptonic ∼
field. If this is correct, the CR electrons should be
γ-rays(Abdo et al. 2011), whereas a recent comparative
accelerated efficiently in the “Bohm diffusion regime’’
study betweenγ-raysandthe interstellarprotonsis con-
(Uchiyama et al. 2007). Additionally, when the cosmic-
sistent with the hadronic origin of γ-rays (Fukui et al.
rays are accelerated, the magnetic fields amplified by
2012;Fukui2013). Itisnotablethatthenumericalsimu-
cosmic-ray streaming instability (Lucek & Bell 2000).
lationsontheshock-cloudinteraction(Inoue et al.2012)
Moreover, different processes of amplifying magnetic
lendsupportforthehadronicγ-rays,sincethelowenergy
fields may be working in during shock-cloud interac-
CR protons producing the GeV γ-rays cannot penetrate
tion (e.g., Giacalone & Jokipii 2007; Inoue et al. 2009,
into the densest part of the cloud cores, leading to the
2012). The amplified magnetic field may intensify the
hard spectrum just presented by Abdo et al. (2011). A
non-thermalX-rays,whereasthesynchrotronlossdueto
similar conclusionis drawnby Zirakashvili & Aharonian
the amplified field may decrease the number of CR elec-
2 Sano et al.
trons via synchrotron loss (e.g., Kishishita et al. 2013).
In order to test such actions, we need a detailed com-
parisonbetweenthe electronspectrumandthe interstel-
larmatte. Cassam-Chena¨ıet al.(2004)presentedspatial
distributionofabsorbingcolumndensity N (X-ray)and
H
photon index Γ obtained by XMM-Newton at angular
resolutions smoothed to 2–30 arcmin. The N (X-ray)
H
distribution shows significant variations over the whole
SNR(0.4 1022cm−2 N (X-ray) 1.1 1022cm−2).
H
× ≤ ≤ ×
The authors carried out an analysis of X-ray absorption
andcomparedtheX-rayswithCO,Hi,andvisualextinc-
tion and showed that the interstellar medium (ISM) is
correlated with the N (X-ray). Additionally, the value
H
of photon index Γ has also shown the strong variation
(1.8 Γ 2.6). This trend was confirmed by a more
≤ ≤
detailed analysis of XMM-Newton data (Hiraga et al.
2005; Acero et al. 2009). Takahashi et al. (2008) and
Tanaka et al. (2008) detected the hard synchrotron X-
rays up to 40 keV from the SNR by using Suzaku.
Fig.1.—SuzakuXISmosaicimageofRXJ1713.7−3946inthe
The spectra are fitted by interstellar absorbed power-
energyband1–5keV(Sanoetal.2013). Thecolorscaleindicates
law model with the exponential rolloff, suggesting CR count rate on a square root scale and is unit of 10−4 counts s−1
electronsclosetotheBohmdiffusionlimit(Tanaka et al. pixel−1. The solidboxes arethose usedfor spectral analysis. We
2008). also show the spectra enclosed by the black solid lines (see also
Figure 2). The dashed boxes correspond to the regions used to
Sano et al. (2010) presented multi-transition CO data
extract the background spectrum. The two point sources, 1WGA
and comparing with the Suzaku X-rays toward the J1714.4−3945and1WGAJ1713.4−3949,arecircledwithmagenta
northwest of RX J1713.7 3946. These authors found andthoseregionswereremovedfromspectralanalysis.
−
thattheX-raysandmolecularclumpsarewell-correlated
at 1 pc scale, but anti-correlated at 0.1 pc scale. The
latter shows a trend as rim-brightened X-rays around 2. OBSERVATIONS AND DATA REDUCTIONS
the dense molecular clumps. Subsequently, Sano et al.
2.1. X-rays
(2013) compared 18 CO clumps and a Hi clump with
their surrounding X-rays and concluded that the trend 2.1.1. Details of the datasets
is commonly found over the whole SNR. Additionally,
We analyzed the X-ray dataset archive obtained
the authors found the strong correlationbetween the X-
by Suzaku (Data Archives and Transmission System;
rayintensityandthemolecularmassinteractingwiththe
DARTS at ISAS/JAXA). This dataset consists of 17
SNR.Theseresults indicatethatthe synchrotronX-rays
pointings taken at 2005 September (SWG; 3 pointings),
arecloselyrelatedtothesurroundingCOgas. Toexplore
2006 September and October (AO1; 10 pointings), 2010
this correlation, we should estimate the physical param-
February (AO4: 4 pointings) and we mainly used 15
eters such as the absorbing column density N (X-ray),
H pointing of ON sources. These data were already an-
photonindexΓandX-rayflux,inangularresolutionhigh
alyzed and published elsewhere (Takahashi et al. 2008;
enough to compare with the ISM distributions. Such
Tanaka et al. 2008; Sano et al. 2013). We used Software
quantitative comparisons with the ISM mark an impor-
HEASoftversion6.11withpipelineprocessingversion2.0
tant step towardunderstanding the spectral variationof
or2.4andwithstandardeventselectioncriteria(cleaned
X-rays and efficient cosmic-ray acceleration.
event files).
Our aim is to compare the spatial distribution of X-
ray properties with that of the ISM clumps. For this 2.1.2. Imaging
aim, we carried out the spectral extraction and model
The X-ray images are those used in Figure 1a of
fitting in the regions gridded at 2–8 arcmin scales over
Sano et al.(2013). Figure1showsXISmosaicimage(1–
theentireSNR.Inthepresentpaper,weshowadetailed
5 keV) in RX J1713.7 3946 (Sano et al. 2013). Color
comparisonofthespatialdistributionamongNH(X-ray), scheme is in a square−-root scale in 10−4 counts s−1
Γ,X-rayflux andISMbothmolecularandatomicgasat
pixel−1 (pixel size is 16.7′′) smoothed with a Gaussian
angularresolutionsragingfrom2arcminto8arcminde-
∼ ′′
kernel with FWHM 45 . This image is subtracted for
pending on the photon statistics. Section 2 gives obser-
∼
non X-ray background (NXB) and corrected for the vi-
vationsanddatareductionofSuzakuX-rays,NANTEN
gnetting effect by XRT (see more details in Sano et al.
CO and ATCA & Parkes Hi. Section 3 consists of three
2013 Section 2.3).
subsections: Section 3.1 presents the typical spectra of
X-rays; Section 3.2 spatial and spectral characterization
2.1.3. Spectroscopy
of the X-rays; and 3.3 a detailed comparison with the
ISM. Section 4 consists of three subsection: in Section The X-ray spectra of RX J1713.7 3946is represented
−
4.1 discusses the spatial variation of absorbing column by the absorbed power-law model (e.g., Koyama et al.
densityisdiscussed;Section4.2therelationshipbetween 1997). This model is for synchrotronX-rayswith photo-
X-ray flux and ISM; and Section 4.2.2 the efficient ac- electric absorption and the model fitting gives three pa-
celeration of cosmic-ray electrons. The conclusions are rameters, absorbing column density N (X-ray), and the
H
given in Section 5. photon index Γ, in addition to the absorption-corrected
3
dicatedbysolidboxes. AboutonethirdofthewholeSNR
is 2′ grid (4 arcmin2) and 80% of that is better than 4′
grid(16arcmin2). Thesevaluesaremuchbetterthanthe
previousstudies(e.g.,Cassam-Chena¨ıet al.2004,70%of
that is worse than 6′ grid (36 arcmin2)). Each spectrum
is binned to include at least 100 counts per individual
energy bin. After the binning the energy ranges a below
0.4keVandabove12keVbinareexcludedinthefitting.
The best fit parameters in the fitting were used to de-
rive absorbing column density N (X-ray), photon index
H
Γ, and absorption-corrected flux in 3–10 keV, F3−10keV,
as shown in Figure 3. If the background spectrum is
changed to that for other three regions, the values of
N (X-ray)onlychangedsystematicallywith10–20%and
H
it has no effect on the global trend in Figure 3.
2.2. CO and Hi
Fig.2.—TypicalX-rayspectraofnorth(black),center(red)and 12CO(J=1–0) and Hi datasets are NANTEN Galac-
south (blue)regions (for theregions definition seeFigure1). The tic Plane Survey (NGPS) and Southern Galactic
solid lines represent the best-fit absorbed power-law model. The
Plane Survey (SGPS) taken with NANTEN, and
lowerpartshowstheresidualsfromthebest-fitmodels. Thenorth
and center spectra can be fitted by the same absorbing column ATCA & Parkes telescopes (Moriguchi et al. 2005;
density (NH(X-ray)∼0.5 × 1022 cm−2) but with different photon McClure-Griffiths et al. 2005), respectively. The angu-
index (north: Γ=2.1±0.1, center: Γ=2.7±0.1). The center and lar resolutionsare HPBW 2.′6 for CO and 2.′2 for Hi,
southspectraarethesamephotonindex(Γ∼2.7)butwithdiffer- ∼ ′ ∼
entabsorbingcolumndensities(center: NH(X-ray)=0.45+−00..0065 × stiiomnislaarntdo tSyupzicaaklurmXsISnHoiPseDfl∼uc2tu.atTiohnesvaerloec(itCyOr)es0o.l6u5-
1022 cm−2,south: NH(X-ray)=0.89+−00..0098 ×1022 cm−2). km s−1 and 0.3 K ch−1 and (Hi) 0.82 km s−1 and 1.9 K
ch−1, respectively.
The velocity integrated intensities of CO and Hi,
X-ray flux. We show typical X-ray spectra in Figure 2. W(CO) and W(Hi), are converted into molecular col-
Weshallexplainthemethodtoderivethe threeparame- umn density N(H2) and atomic column density NH(Hi)
ters above. First, the SNR was divide into 600 regions and the total proton column density is obtained as
of 2′ 2′ grids, where the number of X-r∼ay counts is NH(H2+Hi) = 2 N(H2) + NH(Hi). A relationship
more×than 160 counts per grid in the energy band 1–5 N(H2) (cm−2) =×XCO (cm−2 (K km s−1)−1) W(CO)
keV. For the 600 regions, only the data taken with FI (Kkms−1)isusedwhereX =2.0 1020(cm×−2 (Kkm
CO
CCD (XIS 0,∼2, and 3) were used to derive X-ray spec- s−1)−1)(Bertsch et al.1993). TheH×ilineisgenerallyas-
tra. Inordertocomparewiththe dataoftheinterstellar sumed to be optically thin and a relationship N (Hi) =
H
medium, CO and Hi, the angular resolution was set to 1.823 1018 W(Hi) is used (Dickey & Lockman1990).
the highest value (Suzaku HPD 2′). In this case, the The r×egions×where the Hi is optically thick and hence
∼
signal leaking into/out of neighboring grid cells is less need to apply the correction for self-absorption. Details
than 50 %, because we generated ARFs (Ancillary Re- of the method is given by (Fukui et al. 2012) in Section
sponse Files) with 4 arcmin grid images for each region. 3.3.3.
This signal leaking is not critical in obtaining the large-
scale trend of the photon index, absorbing column den- 3. RESULTS
sity, and fluxes. In addition, the background spectrum
3.1. Typical X-ray spectra
was estimated in the four regions shown in Figure 1 and
the western background spectrum (α =17h10m31s, Figure 2 shows Suzaku XIS 0+2+3 spectra in
J2000
δJ2000 = 39◦44′22.7′′) was used for every region. The the three typical regions, the north (αJ2000, δJ2000)
two prom−inent point sources (1WGA J1714.4 3945and = (17h14m19.95s, 39◦28′31.11′′), the center (α ,
J2000
1WGA J1713.4 3949) in the field are exclu−ded in the δ ) = (17h13m59−.33s, 39◦45′31.49′′), and the south
analysis. Subseq−uently, the spectra of each region taken (αJ2000 , δ ) = (17h12−m56.75s, 40◦00′31.28′′), and
J2000 J2000
with XIS 0, 2, and 3 are summed up and was fit by the the results of the model fitting. T−he lower parts show
absorbedpower-lawmodeltogenerateRMF(Redistribu- residuals in the fitting indicating that the fitting is rea-
tionMatrixFile)byxisrmfgenandARFbyxissimarfgen sonablygood. TheabsorbingcolumndensityN (X-ray)
H
(Ishisaki et al. 2007). is 0.5 1022 cm−2 both in the north and the center,
Next, in order to reduce the statistical relative errors, ∼ ×
but the photon index is different as Γ = 2.1 0.1 and Γ
someneighboringregionsonsourceweresummedupand ±
=2.7 0.1for the northandthe center,respectively. On
the spectra were again fit by the absorbed power-law ±
the other hand, the south has large absorbing column
model. Specifically, we combined two or more 2 arcmin density N (X-ray) = 0.89+0.09 1022 cm−2 and photon
grids so that the relative error of NH(X-ray) becomes index Γ H2.7. These differ−en0c.0e8s×are seen as significantly
30% or less. The combined grids are selected in the ∼
different spectral shapes in Figure 2.
neighboring regions, which have roughly the same val-
ues in N (X-ray).
H 3.2. Spatial and spectral characterization of the X-rays
Figure1showsthefinaldivisionwith305regionsasin-
4 Sano et al.
Fig. 3.— Maps of the best-fit parameters, (a) absorbing column density NH(X-ray), (b) photon index Γ, and (c) absorption-corrected
X-rayflux F3−10keV, for absorbed power-law model. All maps overlaid with smoothed contours of the Suzaku XIS mosaic images. The
contour levels arefrom at 3.9× 10−4 counts s−1 pixel−1 and aresquare-root-spaced up to 23.9× 10−4 counts s−1 pixel−1 with apixel
size of ∼16′.′7. The color schemes in (a–b) and (c) are liner and square-root-scale, respectively. The units are 1022 cm−2 for NH(X-ray)
and10−12 ergcm−2 s−1 forF3−10keV. Thedefinitionofregionnamesisalsoshownin(d). Thecirclecentered at(l,b)=(347◦.3,−0◦.5)
or(αJ2000,δJ2000)=(17h 13m 34s,−39◦ 48′ 17′′),has3pcinradius.
3.2.1. Absorbing column density westandsouth showthe largestN (X-ray)with 1.1–1.4
H
1022 cm−2 (region in red). Therefore, the absorbing
Figure 3a shows absorbing column density NH(X-ray) ×column density varies from 0.4 1022 cm−2 to 1.4 1022
intheindividualregions,wheretheSuzakuXIS1–5keV cm−2 within the SNR. These×values are mostly×con-
intensityisoverlayedascontours. Weaimedatachieving
sistent with the previous studies with XMM-Newton
that the relative error is 30% at maximum at the 90%
(Cassam-Chena¨ıet al. 2004, typical angular resolution
confidence level by tuning the pixel size. Consequently, ′
8), while the angularresolutionandthe sourcecover-
the average relative error and its maximum value are ∼
age are better in the present study.
confined to be 14% and 30%, respectively, at the 90%
Figure 4a shows absorbing column density N (X-ray)
confidence level. This accuracy is high enough to assess H
and visual extinction (A in unit of mag) derived from
thespatialvariationoftheabsorbingcolumndensityand V
the Digitized Sky Survey I (DSS; Dobashi et al. 2005).
photon index as discussed later. We also defined the
The visual extinction is smoothed to the same binning
region name in Figure 3d.
with the X-rays. The values of A are roughly classified
We find that the global distribution of N (X-ray) V
H into the three levels;low values (A < 2 mag in blue) in
showsashell-likestructureinFigure3a. Inthesoutheast V
the center of the SNR, medium values (2 mag < A <
and the center, N (X-ray) is as small as 0.4–0.5 1022 V
cm−2 (regioninblHue). Onthe otherhand,the nort×heast 3 mag in green) in the northeast and northwest regions,
and high values (A > 3 mag in red) in the southwest
and the northwest show medium values of N (X-ray) = V
H and the south. The general trend of A is similar to
0.7–1.0 1022 cm−2 (region in green), and the south- V
×
5
Fig.4.—(a)DistributionofthevisualextinctionAv(Dobashietal.2005). ThegridseparationsarethesameasFigure3. (b)Correlation
plotbetween theabsorbingcolumndensity NH(X-ray)(inunits of 1022 cm−2)andthe visualextinction Av (inunits ofmag). Theerror
barsaregivenata90%confidence levelinNH(X-ray).
that of N (X-ray). Figure 4b shows a correlation plot We also annotated the positions (center of gravity) of
H
betweenabsorbingcolumndensityN (X-ray)andvisual the molecular clumps (crosses) and the Hi clump (cir-
H
extinction,showingagoodcorrelationwithacorrelation cle) defined by Sano et al. (2013). Additionally, we will
coefficient of 0.83. hereinafter refer to the regions between the clumps as
∼
“inter-clump”. Figure 5a shows a trend that the X-rays
3.2.2. Photon index and Flux are enhanced toward the regions with enhanced ISM in
a pc scale. This trend is most significant in the west of
Figures 3b and 3c show the distributions of photon
the SNR, where the ISM is rich. On the other hand,
index Γ and absorption-corrected flux F3−10keV, respec- in a sub-pc scale, we find that each bright spot of X-
tively. Their relative errors and maximum errors are
rays is lying around the CO and Hi clumps except for
6% and 13% in Figure 3b, and 7% and 23% in Fig-
∼ure 3c at the 90% confidence lev∼el, respectively. Like themolecular◦clu′mp′i′nsouthwest(αJ2000=17h12m25.3s,
δ = 39 557.4 ; named as “clump C’’). These re-
absorbing column density, these quantities show signif- J2000 −
sults are consistent with the previous study (Sano et al.
icant spatial variation. In particular, photon index is
2013) andthe presentworkhas made it possible to eval-
largest with Γ 3 (in orange color) toward the center
∼ uate the trend more quantitatively. Figure 6a shows
and the south. On the other hand, regions with a small
photon index (Γ < 2.4) are distributed as islands inside a correlation plot between F3−10keV and the ISM den-
sity; the linear correlation coefficients (hereafter LCC)
the SNR from the north to the southeast and from the
are 0.55 (for the whole), 0.19 (the eastern half, on
southwest to the northwest. Both regions have areas of ∼ ∼
430 arcmin2 and 150 arcmin2, respectively. The theleftsideofαJ2000 17h13m38s)and 0.54(the west-
∼ ∼
a∼bsorption-corrected∼flux F3−10keV is similar to the 1–5 ernhalf,ontherightsideofαJ2000 ∼17h13m38s),respec-
keV X-rays contours, especially at the brightest peaks tively. The LCC 0.55 does not show strong correlation
∼
in the northwest, west, and southwest. We also find an- but the number of sample in the plot 300 indicates a
∼
other strong peak in the north. These flux excesses are positive correlation at a 5% confidence level according
notdueto asystematicerrorandarepossiblyconnected to a T-test, because the value of test statistic ( 11.7)
∼
with the ISM distribution as discussed in Section 3.3. is greater than that of Student’s t distribution ( 1.97)
∼
(e.g., Taylor 1982).
InFigure5b, wenotethatthe regionshavingthe hard
3.3. Comparison with the ISM: The X-ray flux, photon
spectrumareseentowardtheenhancedISMinthewest,
index and the ISM
whereas the regions of the hard spectrum are also found
Overlays of the interstellar gas at V = 20 to 2
LSR toward the diffuse ISM in the eastern half. Most out-
kms−1withabsorption-correctedfluxF3−10keV−andpho- standing are the ISM peaks toward/around the two X-
ton index Γ are shown in Figure 5a and Figure 5b, re-
ray peaks in the northwest and in the southwest. In
spectively, where the ISM is total proton column den-
addition, we find four regions of the hard spectrum in
sity N (H +Hi) including derived from the CO and Hi.
H 2 the southeast where the ISM is diffuse. Figure 6b shows
The velocity range of the ISM associated is derived by
a correlationplot showing that the relationship between
Moriguchi et al. (2005) and is supported by Sano et al.
the ISM and photon index is different between the west
(2013). based on a good correlation between the ISM
andeast. Inthewesternhalf,thespectrumishardwhen
and X-rays enhanced by the interaction with the SNR.
6 Sano et al.
Fig. 5.— Distribution of (a) the absorption-corrected flux F3−10keV and (b) photon index Γ as Figure 3. The contours indicate the
protoncolumndensityNH(H2+Hi)inavelocityrangefrom−20to2kms−1. Thecontourlevelsare0.6,0.75,0.90,1.1,1.3,1.5,1.7,1.9,
and2.1×1022 cm−2. ThemagentacrossesandcirclearecorrespondedtothepositionsofCOandHiclumps(Sanoetal.2013).
theISMisdenseandthespectrumissoftwheretheISM western half, where the ISM is denser than in the
is diffuse (LCC 0.41). In the eastern half, the ISM eastern half.
∼ −
is diffuse and the photon index shows variation (LCC
0.08), where the ISM density is not apparently re- 4. DISCUSSION
∼ −
lated to the variation of the photon index for the whole
4.1. Spatial variation of the absorbing column density
region LCC is estimated to be 0.16. Additionally,
∼ −
we show correlation plots between the photon index Γ The observations with Suzaku XIS have enabled us
andthe absorption-correctedflux F3−10keV in Figure 6c. to estimate detailed distributions of the absorbing col-
LCCs are 0.62 (the whole), 0.53 (the eastern umn density N (X-ray) and photon index Γ in RX
H
∼ − ∼ −
half) and 0.81 (the western half), respectively. The J1713.7 3946 at a low background level and high pho-
∼ − −
histogramstowardthewestandeastregionsshowasimi- ton statics. Here in the discussion, we first discuss the
lartrendtothephotonindexdistributionwithrespected absorbing column density N (X-ray).
H
the ISM density, while the west region has much higher The spatial distribution of N (X-ray) delineates the
H
X-ray fluxes than the east. We shall discuss about these SNR shell as shown in Figure 3a. This distribu-
results later in Section 4.2.2. tion shows a good correspondence with visual extinc-
Wesummarizethemainaspectsofthepresentanalysis tion shown in Figure 4a. The regression in a straight
as follows (Figures 5 and 6); line calculated from a least-squares fitting gives N (X-
H
ray) (cm−2) = 3.0 1021A (magnitude) for the scat-
V
1. In the western half of the SNR, the ISM density ter plot in Figure 4×b. T·he numerical factor is slightly
is high with significant H2 and the X-rays are en- larger than that of the conventionalrelation NH (cm−2)
hanced. In the eastern half of the SNR, the ISM = 2.5 1021A (magnitude) (Jenkins & Savage 1974).
density is low, being dominated by Hi, and the X- × · V
This is understandable if we consider the distance of
raysare weak. The overallcorrelationbetween the
RX J1713.7 3946 is 1 kpc, since the visual extinction
ISM density and the X-rays is however not signif- −
tends to be under-estimated for a distance larger than a
icantly high with a correlation coefficient of 0.55
∼ few 100 pc by the foreground stars. It is not necessar-
(Figure 6a).
ily true that all N (X-ray) is physically associated with
H
the SNR. Here we need to take into account the contri-
2. The smallest X-ray photon index < 2.4 is seen
bution of the local gas between the SNR and the sun
around/towardthe two CO peaks in the the west-
(Figure 7a). Moriguchi et al. (2005) already estimated
ern half, and towardthe regionof low ISM density
the foregroundcomponent N (H +Hi) (Figure 7a).
in the eastern half (Figure 5b). The largest X-ray H,local 2
Figure 7b shows the distribution of absorbing column
photon index around3 is found towardthe central
density[N (X-ray) N (H +Hi)inthe SNR,which
region of the SNR as well as in the southern edge H H,local 2
−
gives a shell-like distribution of N (X-ray) more clearly
(Figure 5b). H
thanFigure3a. The absorptiontowardthe center ofthe
3. The photon index shows a good correlation with SNR is 0.2 1022 cm−2, whereas that towardthe outer
∼ ×
the X-rays and the low photon index is seen to- boundary is shell-like with absorbing column density of
wardregionsofintense X-raysandvice versa(Fig- 0.5–0.8 1022 cm−2. This shellrepresents a cavity wall
∼ ×
ure6c). Thereisanoffsetby0.3inthiscorrelation of the ISM created by the stellar wind of the SNR pro-
betweentheeasternandwesternhalvesoftheSNR genitor. Theinsideofthecavityishighlyevacuatedwith
inthesensethattheX-raysaremoreintenseinthe densitylowerthan 1cm−3 andthedenseclumpsinthe
∼
7
cavitywallhavehigherdensitylike 102–104cm−3 (e.g.,
∼
Sano et al. 2010; Inoue et al. 2012).
We shall discuss the absorption toward the southeast-
rim. In this region Fukui et al. (2012) identified cold
Hi gas that corresponds to the VHE γ-ray shell. The
cold Hi with low spin temperature of 40 K has den-
sity around 100 cm−3, less than 1000∼cm−3 threshold
∼
density for the collisionalexcitation of the CO emission.
The proton column density of the cold Hi without CO
emissionisestimatedtobe 0.5 1022cm−2 fromtheHi
∼ ×
self-absorptioninthesoutheastrimofRXJ1713.7 3946
−
(Fukui et al. 2012). The present absorption in the SNR
calculated from X-rays has also column density of 0.4–
0.5 1022 cm−2 (Figure 7b) which is consistent with the
×
cold Hi.
We compared the absorbing column density [N (X-
H
ray) N (H +Hi)] (Figure 7b) with N (H +Hi),
H local 2 H 2
−
theprotoncolumndensityphysicallyassociatedwiththe
X-ray emitting shell, in two velocity ranges of the inter-
acting gas in Figures 8 and 9. Figures 8a and 8b show
[N (X-ray) N (H +Hi)] overlayed on the ISM
H H local 2
−
proton column density N (H +Hi) at V 20 to
H 2 LSR
9 km s−1 and 9 to 2 km s−1. This co∼mp−arison
− ∼ −
shows that the ISM distribution has generally good cor-
respondencewiththeX-rayabsorbingcolumndensity. It
ishowevertobenotedthattheblue-shiftedISMshowsa
poorer correlationthan that of the red-shifted ISM with
theX-rayabsorbingcolumndensityasshowninFigure9,
scatter plots between the ISM and the X-ray absorption
column density. In Figure 9 we divide the plots into the
red- and blue-shifted ISM. The blue-shifted ISM in the
west shows a poor correlation with the X-ray absorbing
column density of LCC 0.25, while the red-shifted ISM
∼
in the west has a higher correlation of LCC 0.39 than
∼
the blue-shifted ISM (see also Table 1). This indicates
that the red-shifted ISM is locatedon the far-side of the
SNR, and it is in the opposite sense to what is expected
in an expanding motion of the shell-like ISM. The rela-
tive position of the ISM is explicable if the pre-existent
cloud motion is dominant for the dense CO gas instead
ofexpansion,asalreadysuggestedtoexplainthevelocity
distributionoftheHiself-absorptionFukui et al.(2012).
4.2. Relationship between the X-ray flux, the photon
index, and the X-ray absorption/the ISM
4.2.1. Shock-cloud interaction
Inoue et al. (2012) showed by magnetohydrodynamic
(MHD)numericalsimulationsthattheX-rayintensityis
closely correlated with the ISM by the enhanced mag-
netic fieldarounddense clumps due to turbulence in the
shock-cloud interaction. The interaction creates corre-
lation between the ISM and X-rays at a pc scale via
the magnetic field. Observations by Sano et al. (2010,
2013) showed that pc-scale correlation is seen between
the X-rayintensity and the clump mass interactingwith
the SNR blastwaves,as wellas anti-correlationbetween
them in a sub-pc scale due to exclusion of the CR elec-
tronsinthedenseclumps. ThedistributionsoftheX-ray
Fig. 6.—Correlationplots for (a) the absorption-corrected flux
F3−10keV vs. theprotoncolumndensityNH(H2+Hi),(b)thepho- intensityinFigure5aatgridsizesof2–8arcmin(0.6–2.4
ton index Γ vs. the proton column density NH(H2+Hi), and (c) pc) show their interrelation at a pc scale; we see trend
the photon index Γ vs. the proton column density NH(H2+Hi), that the X-raysare enhancedin the westwhere the ISM
respectively. The regions are defined as follow Figure 5: (cid:13) for
is rich, and that the X-rays are depressed in the east
East,✷forWest. Thehistogramsarealsoshownatlowerorright
where the ISM is poor. This is consistent with what
sideofeachplot.
8 Sano et al.
TABLE 1
Summaryof the Correlation Coefficientin Scatter Plots
PhysicalParameters CorrelationCoefficient(LCC)
X-axis Y-axis Wholeregion Easthalf Westhalf Relatedplot
NH(X-ray) AV 0.83 ··· ··· Figure5b
F3−10keV NH(H2+Hi) 0.55 0.16 0.53 Figure6a
Γ NH(H2+Hi) −0.16 −0.09 −0.41 Figure6b
Γ F3−10keV −0.62 −0.53 −0.81 Figure6c
NH(X-ray)−NH,local(H2+Hi) Blue-ShiftedISM 0.25 0.18 0.25 Figure9a
NH(X-ray)−NH,local(H2+Hi) Red-ShiftedISM 0.39 0.45 0.39 Figure9b
Note. — NH(X-ray): absorbing column density, AV: visual extinction, F3−10keV: absorption-corrected X-ray
flux,Γ: photonindex,NH(X-ray)−NH,local(H2+Hi): protoncolumndensity.
Fig. 7.—(a) Distributionof the proton column density inlocal gas NH,local(H2+Hi)estimated byusing the CO and Hidata sets. (b)
Distributionof the absorbing column density [NH(X-ray)−NH,local(H2+Hi)] overlaid with smoothed contours of the SuzakuXIS mosaic
imagesasshowninFigure3. Bothimagesarethesamecolorscale.
is expected in the shock-cloud interaction scheme. The teraction is effective and that turbulence is enhanced
higher dispersion in Figure 6a may be in part ascribed around dense ISM clumps. On the other hand, in the
to anti-correlationat a sub-pc scale, consistent with the eastern half of the SNR where the ISM is poor the DSA
suggestion by Sano et al. (2010, 2013). aloneismainlyworkingwithoutshock-cloudinteraction.
Sano et al.(2013)presentedthattheinnermostcavityof
4.2.2. Efficient cosmic-ray acceleration
3–4 pc radius is surrounded by the ISM shell consisting
Basedonthepresentresultsonthephotonindex(Fig- of the dense clumps and the inter-clump gas. The dense
ure 3b), we discuss the efficient cosmic-ray acceleration clumpshavedensityof102to104cm−3(Sano et al.2010,
in RX J1713.7 3946. As shown by the previous works 2013) and are mainly distributed in the western half to-
−
(Takahashi et al. 2008; Tanaka et al. 2008), the photon wardtheGalacticplane. Theinter-clumpgashasdensity
indexinthe1–10keVrangereflectsrolloffenergyε ofthe < 2 cm−3, as given by the upper limit from no thermal
0
synchrotronX-rays. Thesmall(large)photonindexcor- X-rays (Takahashi et al. 2008), and we shall adopt den-
respondstolarge(small)rolloffenergy. Accordingtothe sity of the inter-clump gas to be 1 cm−3 for discussion.
standard DSA scheme, the rolloff energy of synchrotron We assumethatdensityinthe cavityis 0.25cm−3, the
photonsε0 isgivenasfollowswhenthesynchrotroncool- samewiththatestimatedtowardGumn∼ebulawherethe
ing is effective (Zirakashvili & Aharonian 2007), gas is swept up by the strong stellar winds from ζ Pup
ε =0.55 (v /3000kms−1)2η−1 (keV), (1) (Wallerstein & Silk1971;Gorenstein et al.1974). Inthis
0 × sh scheme, DSA is efficiently taking place in the cavity and
where v is the shock speed and η = B2/δB2 (>1) a the accelerated CRs are injected into the ISM shell.
sh
gyro-factor,the degree of magnetic field fluctuations. In In the western half of the SNR outside the center
particularthelimitofη=1iscalledasBohmlimitcorre- (Figure 3d), where the shock-cloud interaction is work-
spondingtohighlyturbulentconditionsand,accordingly, ing (e.g., Sano et al. 2010; Inoue et al. 2012; Sano et al.
therolloffenergyislikelydeterminedbytheshockspeed 2013),the magneticfieldmaybe amplifiedupto 1mG
∼
and turbulence. as indicated by the short time variation of 1 yr in the
∼
The present results suggest that in the western half synchrotron X-rays by Chandra, where the Bohm limit
of the SNR, where the ISM is rich, the shock-cloud in- (η =1)is agoodapproximation(Uchiyama et al.2007).
9
Fig.8.—Distributionoftheabsorbingcolumndensity[NH(X-ray)−NH,local(H2+Hi)]asFigure7(c)superposedontheprotoncolumn
density NH(H2+Hi)inthe velocity rangefrom(a) −20to −9km s−1 and from(b) −9to2km s−1, respectively. Thecontour levels are
0.4,0.5,0.6,0.7,0.9,1.1,1.3,and1.5×1022 cm−2. Theregions,EastandWest,whereusedforthecorrelationplots(seeFigure9).
Fig.9.— Correlation plot between the absorbing column density [NH(X-ray)−NH,local(H2+Hi)] and the proton column density
NH(H2+Hi) in the velocity range from −20 to −9 km s−1 in (a) and from −9 to 2 km s−1 in (b), respectively. The dashed lines
arethebisectorforeachpanel. Thehistogramsarealsoshownatlowerorrightsideofeachplot. Allplots,thescaleissquare-root.
Ontheotherhand,inthe easternhalftheshockspeedis becomeslargerandthe photonindex smaller asΓ<2.4.
not much decelerated in the lower density. Because the InFigure5b,thecenteroftheSNR(Figure3b)having
shock speed v is proportional to 1/√n and decreases thelargephotonindexΓ>2.8is shownby orangecolor.
sh
inthe denseISM,wherenistheaveragenumberdensity Wesuggestthattheelectronstherehavelowerrolloffen-
of the ISM. The average ISM density in the eastern half ergy due to synchrotron cooling over the last 1000 yrs;
is about 1/4 of that in the western half from the CO/Hi formagnetic fieldof10 µGthe CRelectroncoolingtime
observations and the shock speed v becomes larger by at 10 keV is small as 500 yrs, whereas that at 1 keV
sh
∼
afactorof2. Then, η inthesoutheastbecomeslargerby is largeas 1500yrs,leading to low rolloffenergyin the
afactorof4,buttheterm(theshockspeedv )2 ismore central 3–4∼pc. We also note that such spectral soften-
sh
effective. The trend in the photon index is thus largely ing can be ascribed to the energy loss due to adiabatic
explainedintheDSAschemebyconsideringtheeffectof expansion (e.g., Kishishita et al. 2013).
the ISM gas, and in the both regions the rolloff energy Onewouldexpectthatthephotonindexbecomeseven
10 Sano et al.
larger toward the denser regions, if only v determines to the scale of the ISM distribution, a few ar-
sh
the photon index. Interestingly, the photon index be- cmin,byusingSuzakuarchivaldatawithlowback-
comessmalltowardthe clumpCandthe regionbetween ground.
the two dense clumps, D and L in Figure 5b. It is also
notable that the X-rays are enhanced toward these re- 2. The X-ray flux shows enhancement toward the
gions having small photon indexes. In the shock-cloud dense ISM.This is consistentwith the shock-cloud
interactionscheme,the magneticfieldis amplifiedinthe interaction model by (Inoue et al. 2012) that X-
interactingregion,leadingtohighersynchrotronlossand rays become bright around the dense cloud cores
smaller rolloff energy. This can lead to lower X-ray in- due to turbulence amplification of magnetic field.
tensityandlowerrolloffenergytowardthedenseclumps,
while obviously we need a more elaborate work to quan- 3. Photon index shows a large variation within the
titatively affirm this. It remains to be explored if the SNR from Γ = 2.1–2.9. The photon index
DSA scheme can accommodate the high X-rayintensity, shows smallest values around the dense regions
the high ISM density, and the small photon index con- of cloud cores as well as toward diffuse regions
sistently. Also we here suggest that some additional ac- with no molecular gas. This trend can be de-
celeration mechanism may possibly be working to accel- scribed as a rolloff energy variation of CR elec-
erate CR particles in the west where the ISM is rich. trons(Zirakashvili & Aharonian2007). Wepresent
Such mechanisms possibly include the 2nd order Fermi possible parameters to explain the variation of the
acceleration (Fermi 1949), magnetic reconnection in the rolloff energy under the DSA scheme. The en-
turbulent medium (Hoshino 2012), reverse shock accel- hanced intensity and harder spectra of the X-rays
eration(e.g.,Ellison2001), non-liner effectofDSA (e.g., toward the dense clumps may require additional
Malkov & O’C Drury 2001) and so on. electronaccelerationtowardthedenseclumpspos-
To summarize, we suggestthat the ISM and its distri- sibly via magnetic reconnection or other mecha-
butionhaveasignificantimpactontheschemeofthepar- nism incorporating magnetic turbulence.
ticle acceleration via the shock-cloud interaction. This
suggeststhe tightrelationshipbetweenthe SNRandthe 4. The absorbing column density shows a good cor-
ISM.ThepresentstudyexploredtheCRelectronbehav- relation with the visual extinction. We found that
ior into significant details not achieved in the previous the southeast-rim identified by Hi self-absorption,
works,andpresentedthattheISMdensityaffectssignifi- which shows enhanced VHE γ-rays has a clear
cantlytheCRelectronrolloffenergy. Inparticular,dense counterpart in the X-ray absorption too, lending
molecular clumps excites turbulent in the shock waves, new support to the Hi self-absorption interpreta-
amplifying the magnetic field. The CR electron rolloff tion.
energy is increased in spite of such amplified field, and
we suggest that additional electron acceleration may be
taking place in such turbulent regimes around the dense
clumps. NANTEN2 is an internationalcollaborationof 10 uni-
Innearfuture, CherenkovTelescopeArray(CTA) will versities; Nagoya University, Osaka Prefecture Univer-
provideinformationonthespectraldistributionofγ-rays sity, University of Cologne, University of Bonn, Seoul
over the SNR at 1 arcmin resolution and will allow us National University, University of Chile, University of
to investigate the spectra and acceleration of both the New South Wales, Macquarie University, University of
CR protons and electrons. In addition, the soft X-ray Sydney, and University of ETH Zurich. This work was
spectrometer (SXS) of ASTRO-H will probe yet unde- financially supported by a grant-in-aid for Scientific Re-
tectedthermalX-raysbysensitivehigh-energyresolution search (KAKENHI, No. 2410082, No. 21253003, No.
spectroscopy in RX J1713.7 3946. We can then probe 23403001, No. 22540250, No·. 22244014, No. 23740149,
−
the fraction of the energy of the SNR blast waves used No. 23740154 (T.I.), No. 22740119, and No. 24224005)
for cosmic-ray acceleration. And the hard X-ray imager fromtheMinistryofEducation,Culture,Sports,Science
(HXI) of ASTRO-H will resolve photon index distribu- and Technology of Japan (MEXT). This work was also
tionat10keVorhigher,possiblyallowingustoconstrain financially supported by the Young Research Overseas
electron energy spectrum models by comparing that be- Visits Programfor Vitalizing Brain Circulation (R2211)
low 10 keV (e.g., Yamazaki et al. 2014). These future and the Institutional Program for Young Researcher
instruments will enable more accurate measurements of Overseas Visits (R29) by Japan Society for the Promo-
the efficient cosmic-ray acceleration. tion of Science (JSPS) and by the Mitsubishi Founda-
tionandbythegrant-in-aidforNagoyaUniversityGlobal
5. CONCLUSIONS
COEProgram,“QuestforFundamentalPrinciplesinthe
We summarize the present work as follows. Universe: From Particles to the Solar System and the
Cosmos’’, from MEXT. This research made use of data
1. We have estimated the spatial distribution of obtained from Data ARchives and Transmission System
absorbing column density, photon index, and (DARTS),providedbyCenterforScience-satelliteOper-
absorption-corrected flux (3–10 keV) comparable ation, and Data Archive (C-SODA) at ISAS/JAXA.
REFERENCES
Abdo,A.A.,Ackermann,M.,Ajello,M.,etal.2011,ApJ,734,28 Aharonian,F.A.,Akhperjanian,A.G.,Aye,K.-M.,etal.2004,
Acero,F.,Ballet,J.,Decourchelle, A.,etal.2009,A&A,505,157 Nature,432,75