Magnetically Confined Interstellar Hot Plasma in the Nuclear Bulge of our Galaxy
Shogo Nishiyama, Kazuki Yasui, Tetsuya Nagata, Tatsuhito Yoshikawa, Hideki Uchiyama, Rainer Schödel, Hirofumi Hatano, Shuji Sato, Koji Sugitani, Takuya Suenaga, Jungmi Kwon, Motohide Tamura
aa r X i v : . [ a s t r o - ph . GA ] M a y Magnetically Confined Interstellar Hot Plasmain the Nuclear Bulge of our Galaxy
Shogo Nishiyama , Kazuki Yasui , Tetsuya Nagata , Tatsuhito Yoshikawa , HidekiUchiyama , Rainer Sch¨odel , Hirofumi Hatano , Shuji Sato , Koji Sugitani , TakuyaSuenaga , Jungmi Kwon , and Motohide Tamura ABSTRACT
The origin of the Galactic center diffuse X-ray emission (GCDX) is still under intense inves-tigation. In particular, the interpretation of the hot ( kT ≈ not correlate with thenumber density distribution of an old stellar population traced by near-infrared light, stronglysuggesting a significant contribution of the diffuse interstellar plasma. Contributions of the oldstellar population to the GCDX are implied to be ∼
50 % and ∼
20 % in the Nuclear stellar diskand Nuclear star cluster, respectively. For the Nuclear stellar disk, a scale height of 0 . ◦ ± . ◦ ◦ × ◦ region of our Galaxy,and confirm that the GCDX region is permeated by a large scale, toroidal magnetic field as previ-ously claimed. Together with observed magnetic field strengths close to energy equipartition, thehot plasma could be magnetically confined, reducing the amount of energy required to sustain it. Subject headings:
Galaxy: center — X-rays: ISM — polarization — ISM: magnetic fields
1. INTRODUCTION
In the late 80s, a Japanese X-ray satellite
GINGA revealed the presence of a diffuse andrather uniform 6.7 keV emission from highly ion- National Astronomical Observatory of Japan, Mitaka,Tokyo 181-8588, Japan Department of Astronomy, Kyoto University, Kyoto606-8502, Japan Department of Physics, The University of Tokyo,Bunkyo-ku, Tokyo 113-0033, Japan Instituto de Astrof´ısica de Andaluc´ıa (IAA)-CSIC,18008 Granada, Spain Department of Astrophysics, Nagoya University,Nagoya 464-8602, Japan Graduate School of Natural Sciences, Nagoya City Uni-versity, Nagoya 467-8501, Japan Department of Astronomical Sciences, Graduate Uni-versity for Advanced Studies (Sokendai), Mitaka, Tokyo181-8588, Japan ized, Helium-like ions of iron at the direction of theGalactic center (GC; Koyama et al. 1989). Theline emission and associated continuum compo-nent, called the Galactic center diffuse X-ray emis-sion (GCDX), resembles the Galactic ridge diffuseX-ray emission (GRXE; e.g., Cooke et al. 1969;Koyama et al. 1986) extending more than 100 ◦ along the Galactic plane. For the GRXE, morethan 80 % of the diffuse emission has been claimedto be resolved into point sources (Revnivtsev et al.2009), suggesting faint X-ray point sources in ori-gin.On the other hand, the origin of the GCDX, inparticular its very hot component with a temper-ature of kT ∼ Chan-dra satellite (Muno et al. 2004; Revnivtsev et al.2007). The plasma temperature, represented by1ux ratios of iron emission lines, is systemati-cally higher for the GCDX than for the GRXE(Yamauchi et al. 2009), indicating their differentorigins.Two main ideas have been suggested to accountfor it : a truly diffuse plasma that bathes the emit-ting region (e.g., Koyama et al. 1989); and a su-perposition of a large number of unresolved pointsources as the GRXE is (e.g., Wang et al. 2002).In the later case, candidates are old stellar bi-nary systems such as cataclysmic variables (CVs)and coronally active binaries (ABs; Sazonov et al.2006). So if the hot component originates in thediscrete sources, its spatial distribution should bevery similar to that derived by old stars observablein infrared wavelengths. For this purpose, a stellarmass distribution model constructed from infraredsurface brightness maps (Launhardt et al. 2002)has been used (Muno et al. 2009; Uchiyama et al.2011; Heard & Warwick 2013). However, suchmaps could be subject to the influence of brightstars. The angular resolution in these maps wasonly 0 . ◦
7, so that the stellar density profile in thedirection orthogonal to the Galactic plane hadto be inferred from proxies (dust emission, radioemission from molecular clouds). In addition, un-certainties of the mass model seem to be as highas a factor of two (Launhardt et al. 2002).We have constructed a stellar number density map of the GC region from new near-infrared(NIR) observations with more than 1,000 timeshigher spatial resolution (Yasui et al. in prepara-tion), which enables us to directly compare thestellar distribution with GCDX. In this letter ,we summarize the NIR imaging observations andtheir results, and provide additional evidence forthe hypothesis that the GCDX arises from a trulydiffuse hot plasma. We also show results of ourrecent polarimetric observations. The results pro-vide strong evidence for a large-scale toroidal mag-netic field configuration which could confine thehot plasma magnetically.
2. Observations and Data Analysis
The central region of our Galaxy, | l | . . ◦ | b | . . ◦ ×
280 pc at 8 kpc from the Sun), was observedfrom 2002 to 2004 using the NIR camera SIRIUS(Nagashima et al. 1999; Nagayama et al. 2003) on the 1.4 m telescope IRSF. SIRIUS provides J (1.25 µ m), H (1.63 µ m), and K S (2.14 µ m) images si-multaneously. The averages of the 10 σ limitingmagnitudes are H = 16 . K S = 15 .
6. Wedo not use the J -band data due to severe inter-stellar extinction. Further details are given inNishiyama et al. (2006).The stellar number density map is constructedas follows: At first, an H and K S color magni-tude diagram (CMD) is constructed for each sub-field of 20 ′ × ′ , and foreground sources with theirblue H − K S color are removed. The typical colorof stars in the GC is H − K S > .
0, and colorcuts to remove the foreground sources are 0.3 -1.1. We carry out an extinction correction foreach star using the observed H − K S color, themean intrinsic color of ( H − K S ) ≈ .
20 (con-sidering the limiting magnitudes and the Galacticmodel by Wainscoat et al. 1992), and an interstel-lar extinction law, A ( K S ) = 1 . × { ( H − K S ) − ( H − K S ) } (Nishiyama et al. 2006). Here we ob-tain an extinction-corrected K S -band magnitude, K S, , and the amount of interstellar extinction foreach star. -3-2-1 0 1 2 3 l (deg)-1.5-1-0.5 0 0.5 1 1.5 b ( deg ) N u m be r o f S t a r s ( a r c m i n - ) Fig. 1.— The stellar number density map of thecentral 6 ◦ × ◦ region of our Galaxy. Stars with K S, < . < . ′ ,and for the outside of 20 ′ , respectively. Severallow-density regions are seen in the NB, and arenot used in the following analysis.We then construct a stellar number densitymap using stars with K S, < . ′ × ′ field, source confusionis so severe that a different magnitude limit of K S, < .
0, and a conversion factor derived bythe ratio of the number of stars with K S, < . K S, < . K S -2and luminosity functions constructed with theextinction-corrected stars and a Galactic model(Wainscoat et al. 1992), these magnitude limitsare determined so that completeness at the limitsis almost 100 %. We make completeness correc-tions with recovery rates determined by adding ar-tificial K S = 12 . ∼
96 %in average (Hatano et al. 2013). For more de-tail, see Yasui et al. (in preparation). Also, atthe very center ( < ′ ), we have used images ob-tained with an 8-m telescope VLT and ISAAC(Nishiyama & Sch¨odel 2013).To determine a large-scale interstellar MF con-figuration, we have carried out NIR polarimetricobservations using IRSF and a NIR polarimetricimager SIRPOL (Kandori et al. 2006), from 2006to 2010. We have extended the survey region from | l | . . ◦ | b | . . ◦
0; Nishiyama et al. 2010)to | l | . . ◦
5, which covers almost the whole regiondominated by GCDX. Comparing the polarizationbetween stars distributed further and closer side inthe GC, we obtain polarization originating frommagnetically aligned dust grains in the GC (formore detail, see Nishiyama et al. 2009, 2010). Thepolarized angle traces the GC’s MF direction pro-jected onto the sky.
3. Results and Discussion3.1. Stellar Number Density Profiles
A disk-like structure is seen in the stellar num-ber density map (Fig. 1). This is known as theNuclear bulge (NB), which consists of the Nuclearstellar disk (NSD) and the Nuclear stellar clus-ter (NSC; Launhardt et al. 2002). Here our ob-servations clearly reveal morphology of the NBon a large scale, with much higher spatial reso-lution than previous studies. The NB has a sym-metric, disk-like structure with a scale height of0 . ◦ ± . ◦
02 (Fig. 2), although several low-densityregions are seen. Those are very dense molecu-lar clouds in front of/inside the NB. A true stellarnumber density is difficult to be derived in theseregions, and thus they are not used in the followinganalysis (see Yasui et al. in preparation).The longitudinal and latitudinal profiles ofthe 6.7 keV line emission measured by
Suzaku (Koyama et al. 2007; Uchiyama et al. 2011) clearlyshow an excess at the NB region over the stellar
10 100 0.01 0.1 1 N u m be r o f S t a r s ( / a r c m i n ) |l * | (deg)
10 100 0.01 0.1 1 N u m be r o f S t a r s ( / a r c m i n ) |b * | (deg) Fig. 2.— Top: Longitudinal profile of the stellardensity distribution after a completeness correc-tion. Only data points at l ∗ < l ∗ = 0 . ◦ b ∗ are plotted as green (cyan)and red (pink) marks, respectively. The cyan andpink marks are not used for the fitting because thenumber densities are underestimated in the corre-sponding regions due to strong line-of-sight extinc-tion. Data points at | b ∗ | < . ◦ . ◦
3. The NSD componenthas a scale height of 0 . ◦ ± . ◦
02. The numberdensities are calculated in rectangular bins with asize of 1 ′ ( l ) × ′ ( b ) for the longitudinal profile at b ∗ = 0 ◦ , and 2 ′ ( l ) × ′ ( b ) for the latitudinal profileat l ∗ = 0 ◦ .3umber density profiles (Fig. 3). The two profilesare overplotted and scaled to have the same valuesat 1 . ◦ < | l ∗ | < . ◦
8, i.e., in a region outside of theNB [ l ∗ and b ∗ denote the angular distance fromSgr A* along the Galactic longitude and latitude,respectively, and ( l ∗ , b ∗ ) = ( l + 0 . ◦ , b + 0 . ◦ − . ◦ ≤ l ∗ ≤ − . ◦ ∝ θ − α ,where θ is angular offset from Sgr A*, gives α star = 0 . ± .
03 for the stellar number den-sity. This is different both from 0 . ± .
02 forthe 6.7 keV profile in the same range, and from0 . +0 . − . for the integrated emission of Fe 6.7 and6.9 keV lines (Heard & Warwick 2013).The majority of faint X-ray ( L −
10 keV < erg s − ) sources which have not been re-solved but contribute to the GCDX are mostlikely to be old binary systems (Sazonov et al.2006). Using the synthetic CMD computation(Aparicio & Gallart 2004), and a constant starformation history during 13 Gyr for the NSD(Figer et al. 2004), we have confirmed that about75 % of the stars with K S, < . . ◦ < | l ∗ | < . ◦
8, the ratios are ∼ . ∼ N all , to the number of stars with K < . N K< . . This ratio, R ≡ N all /N K< . , repre-sents the ratio of the theoretically expected totalnumber of stars to the number of stars detectedin our observations. The star formation histo-ries used here are: a burst star formation from
10 100 -3-2-1 0 1 2 10 -7 -6 N u m be r o f S t a r s ( / a r c m i n ) F e XXV K α ( pho t on s / s c m a r c m i n ) l * (deg) 10 100 -1-0.5 0 0.5 1 10 -7 -6 N u m be r o f S t a r s ( / a r c m i n ) F e XXV K α ( pho t on s / s c m a r c m i n ) b * (deg)l=-0.17 Fig. 3.— Longitudinal (top) and latitudinal (bot-tom) profiles of the stellar number density aftera completeness correction (red crosses). Over-plotted are the 6.7 keV-emission profiles (blue x;Koyama et al. 2007; Uchiyama et al. 2011). Theregion outside the NB, 1 . ◦ ≤| l ∗ |≤ . ◦
8, is usedto scale the the 6.7 keV emission profile to havethe same value as the stellar number density. Thenumber density is calculated in the same rectan-gles as those used in Uchiyama et al. (2011), witha size of 0 . ◦ l ) × . ◦ b ) for the longitudinal profile,and 0 . ◦ l ) × . ◦ b ) at the position of l = − . ◦ F e XXV I n t en s i t y / N u m be r o f S t a r s ( . × - pho t on s s - c m - a r c m i n - / a r c m i n - ) l * (deg) 0.5 1 1.5 2 2.5 -1-0.5 0 0.5 1 F e XXV I n t en s i t y / N u m be r o f S t a r s ( . × - pho t on s s - c m - a r c m i n - / a r c m i n ) b * (deg)l=-0.17 Fig. 4.— Longitudinal (top) and latitudinal (bot-tom) profiles of the ratios of the 6.7 keV emissionto the stellar number density, scaled to be unityat the position for normalization, 1 . ◦ ≤| l ∗ |≤ . ◦ R to be unity for the GB,we obtain R GB : R NSD : R NSC ≈ . . / . × . ∼ . / × . ∼ . ∼ / byCVs and ABs , rather than a higher emissivity. The most puzzling aspects of the GCDX isits high temperature. Since the kT ≈ without the con-finement of the plasma. One idea to addressthis energetics problem is the confinement of theplasma by magnetic fields (MFs; Makishima 1994;Tanuma et al. 1999). If a large-scale toroidal5F is developed and sustained, and the MF isstrong enough for nearly energy equipartitionwith the plasma, the GCDX could be almost con-fined within the NB. However, the large scale MFconfiguration was thought to be predominantlyvertical, suggesting that the magnetic confine-ment does not work well, although observationsof the MF in the NB have been limited to theregion in dense molecular clouds (Novak et al.2000, 2003; Chuss et al. 2003) and very thin,non-thermal radio filaments (Tsuboi et al. 1986;Yusef-Zadeh et al. 1997; Lang et al. 1999). Recentobservations have revealed that NIR and wide-field polarimetry offers a promising tool to tracea large-scale MF, and a troidal configuration nearthe Galactic plane has been claimed in the GC(Nishiyama et al. 2010).Fig. 5.— Polarimetry results covering 3 . ◦ × . ◦ . ′ . ′ σ ) and thin bars (detected with 2 − σ ).The obtained polarization map (Fig. 5) sug-gests a large-scale toroidal MF configuration in theNB. The histogram (Fig. 6) of the MF directionsat | b | < . ◦ ◦ which is thedirection parallel to the Galactic plane. On theother hand, at high Galactic latitude ( | b | & . ◦ b ∼ . ◦ − . ◦
4, is in good agreementwith the scale height of the 6.7 keV emission, 0 . ◦ Position Angle [degree] N u m be r o f F i e l d s all| b | < 0.4 o -| b | > 0.4 o Fig. 6.— Histograms of the magnetic field direc-tion at | b |≤ . ◦ | b | > . ◦ ◦ ), while the blue one has a peak at ∼ ◦ ,almost perpendicular to the plane.The determination of the MF strength is stillquite difficult in this region, but it seems toconverge to the value of 50 . B [ µ G] . ∼ . − − ) reaches nearly equipartitionwith those of diffuse hot plasma ( ∼ . − )and gas turbulence (see Fig. 4 in Crocker et al.2010). This suggests that MFs provide significantpressure support against the diffusion of the hotplasma.If the plasma were not supported, it wouldbe rushing out of the Galactic plane verticallyas a galactic wind. The escape velocity, typ-ically several hundred km s − , is smaller thanthe sound speed of the 7 keV hot plasma of ∼ ,
400 km s − . Assuming the gas flows out fromthe X-ray emitting region at the sound speed, the6scape timescale is ∼ × yr (Belmont et al.2005). This requires a huge energy input to sus-tain the hot plasma; e.g., an unreasonably high su-pernova rate of ∼ × − yr − (Uchiyama et al.2013, but note that kT ≈ − yr(Muno et al. 2004), several orders of magnitudelonger than the escape timescale. This would re-duce the required energy input by several orders ofmagnitude and thus relax the energetics problem.There is no widely accepted mechanism to heatthe plasma to kT ≈
4. Summary
We have used imaging and polarimetric datasets of the GC region to investigate the origin ofthe GCDX. We have constructed a stellar numberdensity map, and compared its longitudinal andlatitudinal profiles with those of the 6.7 keV emis-sion. We have estimated that the contributionsof the old stellar population to the GCDX at theNSD and NSC are ∼
50 % and ∼
20 %, respec-tively. Our findings support the notion that theGCDX is not only caused by a population of un-resolved point sources but must also stem from ahot interstellar plasma component. Polarimetricobservations reveal a large scale toroidal magneticfield configuration which allows a magnetic con-finement of the hot plasma.This work was supported by JSPS KAKENHIGrant numbers 23840044, 22000005, 25707012,Grant-in-Aid for the JSPS Fellows 20 · REFERENCES
Aparicio, A., & Gallart, C. 2004, AJ, 128, 1465Bastian, N., Covey, K. R., & Meyer, M. R. 2010,ARA&A, 48, 339Belmont, R., Tagger, M., Muno, M., Morris, M.,& Cowley, S. 2005, ApJ, 631, L53Carretti, E., Crocker, R. M., Staveley-Smith, L.,et al. 2013, Nature, 493, 66Chuss, D. T., Davidson, J. A., Dotson, J. L., et al.2003, ApJ, 599, 1116Cooke, B. A., Griffiths, R. E., & Pounds, K. A.1969, Nature, 224, 134Crocker, R. M. 2012, MNRAS, 423, 3512Crocker, R. M., Jones, D. I., Aharonian, F., et al.2011, MNRAS, 413, 763Crocker, R. M., Jones, D. I., Melia, F., Ott, J., &Protheroe, R. J. 2010, Nature, 463, 65Ferri`ere, K. 2011, in Astronomical Society of thePacific Conference Series, Vol. 439, The Galac-tic Center: a Window to the Nuclear Environ-ment of Disk Galaxies, ed. M. R. Morris, Q. D.Wang, & F. Yuan, 39Figer, D. F., Kim, S. S., Morris, M., et al. 1999,ApJ, 525, 750Figer, D. F., Rich, R. M., Kim, S. S., Morris, M.,& Serabyn, E. 2004, ApJ, 601, 319Hatano, H., Nishiyama, S., Kurita, M., et al. 2013,AJ, 145, 105Heard, V., & Warwick, R. S. 2013, MNRAS, 428,3462Kandori, R., Kusakabe, N., Tamura, M., et al.2006, in Society of Photo-Optical Instrumenta-tion Engineers (SPIE) Conference Series, Vol.6269, Society of Photo-Optical InstrumentationEngineers (SPIE) Conference Series7oyama, K., Awaki, H., Kunieda, H., Takano, S.,& Tawara, Y. 1989, Nature, 339, 603Koyama, K., Maeda, Y., Sonobe, T., et al. 1996,PASJ, 48, 249Koyama, K., Makishima, K., Tanaka, Y., &Tsunemi, H. 1986, PASJ, 38, 121Koyama, K., Takikawa, Y., Hyodo, Y., et al. 2009,PASJ, 61, 255Koyama, K., Hyodo, Y., Inui, T., et al. 2007,PASJ, 59, 245Lang, C. C., Morris, M., & Echevarria, L. 1999,ApJ, 526, 727Launhardt, R., Zylka, R., & Mezger, P. G. 2002,A&A, 384, 112Makishima, K. 1994, in New Horizon of X-Ray Astronomy. First Results from ASCA, ed.F. Makino & T. Ohashi, 171Muno, M. P., Baganoff, F. K., Bautz, M. W., et al.2004, ApJ, 613, 326Muno, M. P., Bauer, F. E., Baganoff, F. K., et al.2009, ApJS, 181, 110Nagashima, C., Nagayama, T., Nakajima, Y.,et al. 1999, in Star Formation 1999, ed.T. Nakamoto, 397–398Nagayama, T., Nagashima, C., Nakajima, Y.,et al. 2003, in Society of Photo-Optical Instru-mentation Engineers (SPIE) Conference Series,Vol. 4841, Society of Photo-Optical Instrumen-tation Engineers (SPIE) Conference Series, ed.M. Iye & A. F. M. Moorwood, 459–464Nishiyama, S., & Sch¨odel, R. 2013, A&A, 549, A57Nishiyama, S., Nagata, T., Kusakabe, N., et al.2006, ApJ, 638, 839Nishiyama, S., Tamura, M., Hatano, H., et al.2009, ApJ, 690, 1648Nishiyama, S., Hatano, H., Tamura, M., et al.2010, ApJ, 722, L23Nobukawa, M., Ryu, S. G., Nakashima, S., et al.2012, in American Institute of Physics Confer-ence Series, Vol. 1427, American Institute ofPhysics Conference Series, ed. R. Petre, K. Mit-suda, & L. Angelini, 209–216 Novak, G., Dotson, J. L., Dowell, C. D., et al.2000, ApJ, 529, 241Novak, G., Chuss, D. T., Renbarger, T., et al.2003, ApJ, 583, L83Pfuhl, O., Fritz, T. K., Zilka, M., et al. 2011, ApJ,741, 108Revnivtsev, M., Sazonov, S., Churazov, E., et al.2009, Nature, 458, 1142Revnivtsev, M., Vikhlinin, A., & Sazonov, S. 2007,A&A, 473, 857Sazonov, S., Revnivtsev, M., Gilfanov, M., Chu-razov, E., & Sunyaev, R. 2006, A&A, 450, 117Tanuma, S., Yokoyama, T., Kudoh, T., et al. 1999,PASJ, 51, 161Tohline, J. E. 2002, ARA&A, 40, 349Tsuboi, M., Inoue, M., Handa, T., et al. 1986, AJ,92, 818Uchida, Y., Sofue, Y., & Shibata, K. 1985, Nature,317, 699Uchiyama, H., Nobukawa, M., Tsuru, T., Koyama,K., & Matsumoto, H. 2011, PASJ, 63, 903Uchiyama, H., Nobukawa, M., Tsuru, T. G., &Koyama, K. 2013, PASJ, 65, 19Wainscoat, R. J., Cohen, M., Volk, K., Walker,H. J., & Schwartz, D. E. 1992, ApJS, 83, 111Wang, Q. D., Gotthelf, E. V., & Lang, C. C. 2002,Nature, 415, 148Yamauchi, S., Ebisawa, K., Tanaka, Y., et al.2009, PASJ, 61, 225Yusef-Zadeh, F., Wardle, M., & Parastaran, P.1997, ApJ, 475, L119Zoccali, M., Renzini, A., Ortolani, S., et al. 2003,A&A, 399, 931