Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

High Resolution Imaging of the Magnetic Field in the central parsec of the Galaxy

High Resolution Imaging of the Magnetic Field in the central parsec of the Galaxy We discuss a high resolution (FWHM 0:45 arcsec) image of the emissive polar- ization from warm dust in the minispiral in the Galactic Centre and discuss the implications for the magnetic eld in the dusty laments. The image was ob- tained at a wavelength of 12.5 m with the CanariCam multimode mid-infrared imager on the Gran Telescopio Canarias. It con rms the results obtained from previous observations but also reveals new details of the polarization structures. In particular, we identify regions of coherent magnetic eld emission at position angles of  45 to the predominantly north{south run of eld lines in the North- ern Arm which may be related to orbital motions inclined to the general ow of the Northern Arm. The luminous stars that have been identi ed as bow-shock sources in the Northern Arm do not disrupt or dilute the eld but are linked by a coherent eld structure, implying that the winds from these objects may push and compress the eld but do not overwhelm it. The magnetic eld in the the low surface brightness regions in the East-West Bar to the south of SgrA* lies along the Bar, but the brighter regions generally have di erent polariza- tion position angles, suggesting that they are distinct structures. In the region of the Northern Arm sampled here, there is only a weak correlation between the intensity of the emission and the degree of polarization. This is consistent with saturated grain alignment where the degree of polarization depends on geometric e ects, including the angle of inclination of the eld to the line of sight and superposition of laments with di erent eld directions, rather than the alignment eciency. Keywords: Galactic Centre, dust, extinction, magnetic elds, polarization Preprint submitted to Elsevier August 1, 2018 arXiv:1807.11744v1 [astro-ph.GA] 31 Jul 2018 1. Introduction The central parsec of the Milky Way is a complex region hosting a range of unusual phenomena, including a supermassive black hole of mass  4 10 M , a young stellar cluster with numerous luminous mass-losing stars together with molecular and ionized gas laments mixed with dust (e.g. Genzel, Eisenhauer & Gillessen 2010). At mid-infrared wavelengths, the most prominent structure is the mini-spiral of gas and dust that comprises the Northern Arm and East- West Bar which lie within the 1.5{5 pc circumnuclear disk of molecular material around the Galactic Centre. The mid-IR emission arises from dust grains heated to temperatures of 180{250 K (e.g. Smith, Aitken & Roche 1990; Gezari et al. 1985). The derived dust temperature is fairly uniform in the bright laments but there are local peaks close to the Northern Arm out ow sources. Polarimetric measurements have revealed coherent polarization structures in the minispiral (Aitken et al., 1986, 1991, 1998; Glasse, Aitken & Roche, 2003), which provide information on the magnetic eld in the laments. The magnetic eld appears to be quite ordered and strong in the Northern Arm, with estimated eld strength of approximately 2 mGauss through indirect arguments, but much less ordered in the East-West Bar. On the assumption that the dust grains have a net align- ment to the eld lines, determination of the polarization allows us to estimate the eld directions in the plane of the sky. Aitken et al. (1998) have argued that the degree of polarization may be saturated in the Northern Arm, in which case, varations in the percentage polarization are related to the inclination of the eld with respect to the plane on the sky. Combination of polarimetric observations with radial velocities of the ionized gas then provides information on the 3-D structure of the laments. Here we discuss new higher resolution imaging polarimetry of the southern part of the Northern Arm and brightest region of the East-West Bar of the minispiral and the detailed correspondence between the polarization structures and other constituents in the central parsec. 2. Polarimetry of the Galactic Centre The centre of the Galaxy is viewed through an 8 kpc long column through the interstellar medium. This renders it invisible at short wavelengths, but the e ects of extinction are much reduced at infrared wavelengths. It appears that the extinction over the central parsec is relatively uniform and characterised by a fairly constant extinction with an optical depth at the peak of the silicate ab- sorption feature,   3:6 (e.g. Roche & Aitken 1985). Spectropolarimetry 9:7m of the compact mid-IR sources in the minispiral has revealed that the polariza- tion signatures can be interpreted by an intrinsic emissive polarization signature overlaid by a constant interstellar polarization component (Aitken et al., 1986). The polarization image presented here was measured at 12.5 m where the emissive component is large while the interstellar absorptive component, which has a pro le similar to the silicate absorption band, has decreased substantially from its peak near 10 m. The interstellar absorptive polarization is discussed 2 Absorptive Emissive Figure 1: Measured infrared spectral dependence of polarization towards IRS1, the brightest compact source in the minispiral in the mid-IR. From the top: Intensity, Percentage Polar- ization and Position Angle. Note the change in position angle with increasing wavelength from the near- to the mid-IR re ecting the change from domination by interstellar absorptive polarization to intrinsic emissive polarization. Note also the rotation in position angle across the 10m silicate band. At short infrared wavelengths and near the peak of the silicate ab- sorption feature, interstellar polarization dominates, but in the wings of the silicate band near 8 and 12 m, the emissive component is dominant. This behaviour is modelled by an emissive component at PA = 120 deg overlaid by an interstellar absorptive component at 0 deg. The observations are made in apertures of  4 arcsec. Near-IR data are from Bailey, Hough & Axon (1984); Knacke & Capps (1977) while the 8-13 and 17-22 m spectropolarimetric data are from Smith et al. 2000. 1996ASPC...97..551R in detail by Aitken et al. (1991, 1998) who compare estimates of the interstellar polarization from mid-infrared spectro- and imaging- polarimetry and conclude that all of the results are consistent with a constant absorptive polarization of 1.8% at PA= 5  5 deg at 12.5 m arising from cool aligned grains in the ISM. The observed spectral dependence of the infrared polarization measured towards IRS1 is shown in Fig. 1, where the changes in position angle reveal the way the contributions from the emissive and absorptive components change with wavelength. The detailed decomposition of the emissive and absorptive components between 8-13 m can be found in Smith et al. (2000). Because of dissipative forces in grains rotating in a magnetic eld, non- spherical dust particles will tend to align so that the grain spin axis is parallel to the maximum moment of inertia. The grain spin axis will precess around the magnetic eld, providing a net alignment of an ensemble of dust grains, so that the long axis of a grain is preferentially aligned to be perpendicular to the magnetic B eld (see e.g. Andersson, Lazarian & Vaillancourt 2015 for a review). This results in polarized emission with the position angle normal to the B eld, so that when the emissive polarization vectors are rotated by 90 , they trace out the component of the eld directions in the plane of the sky. 3. Observations The observations were made with the Canaricam mid-IR imager mounted on the Nasmyth platform of the 10.4-m Gran Telescopio Canarias. The data were obtained in the polarimetric mode wherein a rotating cryogenic half wave plate is used in conjunction with a focal plane mask and a Wollaston prism to produce simultaneous images of the e- and o-rays in three slots, each 2 x 20 arcsec (Packham, Hough & Telesco, 2005). The telescope was stepped by 1 arcsec in Right Ascension after each exposure to provide full sampling and contiguous polarization maps of the minispiral. The Galactic Centre only rises above 2 airmass for short periods above La Palma, and the delivered image quality is seeing-limited at FWHM = 0.45 arcsec. The polarization image was produced by coadding three sets of exposures which were aligned and corrected for the waveplate eciency and for instrumental polarization induced by the o - axis re ection in the Nasmyth mirror according to the prescriptions provided on the GTC/Canaricam website . The instrumental polarization induced by the tertiary mirror has been measured at 0.6% with the position angle dependent on the instrument rotator angle. Observations of unpolarized stars taken with Canaricam (though not on the same nights as the Galactic Centre) indicate residual instrumental polarization < 0:2%. Finally, the image was corrected for interstellar polarization by taking the measured polarization of the isolated luminous supergiant star, IRS3, which has a constant polarization position angle across the silicate band, to be representative of the interstellar component only (see the discussion in Aitken et al. 1998). The measured polarization of IRS3 see http://www.gtc.iac.es/instruments/canaricam 4 in the current data is 1.7% at PA 10 deg and we have used this value to correct for interstellar polarization. On the assumption that IRS3 is intrinsically unpolarised, this procedure corrects for any residual instrumental polarization as well as the interstellar absorptive component. The plate scale of Canaricam is 0.08 arcsec pixel . The data presented here have been smoothed to 0.24 arcsec pixel to improve the signal to noise ratio per point while preserving full sampling. For more details of the data analysis and processing see Roche et al. (2018). 4. Polarization image The emissive polarization map is shown in Fig. 2. In the image, the vectors have been rotated by 90 degrees, so that they point along the magnetic eld directions. It is clear that, as found previously, (e.g. Glasse, Aitken & Roche 2003), the Northern Arm, which is a coherent structure running from the north- east of the image before becoming indistinct south of SgrA* (e.g. Irons, Lacy & Richter 2012) contains substantial regions of coherent vectors and that the bright compact sources neither dilute nor disturb the local eld con gurations. This is seen more clearly and at higher resolution in the current image, where a run of vectors is seen linking the bright objects, IRS5, IRS10, IRS1 and IRS21. These objects have been identi ed as luminous stars with substantial out ows which produce bow shocks where the stellar winds impact on the Northern Arm laments (Tanner et al., 2005; Sanchez-Bermudez et al., 2014), and produce local temperature peaks (Smith, Aitken & Roche 1990). The polarization at these positions closely follows that in the adjacent laments indicating that the out ow sources are embedded in the magnetic elds and that while the out ow energy may compress and push the B elds, it does not disrupt them. The coherent vectors linking the bright embedded out ow sources suggests that these objects have played a role in pushing the magnetic ux tube. In addition to this coherent structure linking the embedded objects and the general N{S vectors following the larger scale structure of the Northern Arm near the top of the image, two regions of coherent vectors crossing the Arm at PA  45 can be identi ed to the east and south-east of IRS1. The rst of these appears to converge or merge with the vector ow immediately south of IRS1 and continue to IRS21; the second becomes indistinct where it meets a region of N{S vectors. These B eld structures may indicate regions where gas is owing across the Northern Arm and in fact they provide a qualitative match to some of the Keplerian orbits proposed by Paumard, Maillard & Morris (2004) and may indicate regions of inward gas ow. Some grain alignment mechanisms, such as radiative torques acting on grains might be expected to lead to greater alignment and higher polarization in regions with high radiation uxes (e.g. Andersson, Lazarian & Vaillancourt 2015). To further investigate this, the percentage polarization is plotted against the pixel intensity in the region of the Northern Arm to the north and east of IRS21 in Fig. 3. Regions below the threshold for reliable polarization estimation (1 Jy arcsec ) are excluded. The intensity varies by a factor of >30 in this 5 IRS5 10% IRS10 IRS3 IRS1 IRS21 IRS6 IRS2 -5 10 5 0 -5 RA Offset (arcsec) Figure 2: Polarization vectors superimposed upon a greyscale of the 12.5 m intensity image. The vectors have been rotated by 90 deg so that they point along the magnetic eld directions. Coordinates are given as o sets in arcsec from the position of SgrA* which is marked by the cross. The positions of some of the bright compact objects are indicated. The lowest contour 2 2 is at 1 Jy arcsec with further contours at 3, 6 and 9 Jy arcsec . The image size is indicated by the circle at the lower right. Hand drawn lines indicate the run of coherent polarization vectors linking the embedded bow shock sources IRS5, IRS10, IRS1 and IRS21, while the lines to the south and east of IRS1 indicate the regions that cross the Northern Arm at angles of  45 . In the East-West Bar, which runs from the south-east corner of the image to IRS6, the regions of lower intensity have polarization vectors pointing along the bar while the polarization in the brighter regions appear to arise from distinct eld con gurations. 6 15 1 10 Figure 3: Polarization plotted against intensity in the region of the Northern Arm to the north and east of IRS21. The intensity is plotted logarithmically to spread out the large number of points at the lower intensity levels. region, peaking on IRS1, while the polarization ranges up to a maximum of 14%. Such a plot has to be interpreted with caution as only the component of polarization in the plane of the sky is measured here. Furthermore, there are places where local minima appear to arise from the superposition of di erently oriented laments, and the distribution of the sources of luminosity and the 3- dimensional structure of the laments will all a ect the measured polarization. 0:23 The best t regression line is P(%)= 4 S , where S is the ux density in Jy, but the correlation coecient is very low with R = 0.10. Inspection of Fig. 3 suggests at most a modest increase in polarization with intensity suggesting that the e ect expected from radiative torque alignment is modest in the central region of the Milky Way. However, Aitken et al. (1998) proposed that the grain alignment in the Northern Arm may be saturated, in which case variations in the degree of polarization along the Northern Arm re ect changes in the orientation of the eld with respect to the line of sight rather than the degree of alignment. As several alignment mechanisms, including radiative torques (e.g. Hoang & Lazarian 2016) may be able to produce saturated alignment in grains with iron inclusions, this may not provide a robust test of the alignment mechanism. At rst sight, the B eld structure in the East-West Bar seems to be much less ordered than that in the Northern Arm. Inspection of the image shows that in the regions with lower emission intensity, the eld lies predominantly along the Bar with PA  130 , but in contrast to the Northern Arm, the brighter 7 regions in the Bar have polarization levels and position angles that diverge from the local di use emission. This is especially apparent south of IRS2, where the polarization vectors are approximately north-south, and quite distinct from the underlying di use vector eld which lies along the Bar. Similarly, near IRS6 at the western end of the Bar, there is a region where the polarization vectors have PA  45 close to the junction with the western arc. It seems that the bright regions in the Bar are not embedded in the general di use B eld but are regions with distinct properties. The region south of IRS2 shows strongly blue-shifted [Ne II] emission (Irons, Lacy & Richter, 2012) and has been considered as the southern extension of the Northern Arm where it re- emerges after curling below SgrA*; the polarization vectors appear to be be consistent with that interpretation. Similarly, the region near IRS6 is where the western arc connects to or is superimposed on the Bar, and again we might expect distinct B eld properties here. 5. Conclusions The 12.5 m polarization image presented here reveals new details of the magnetic eld in the central parsec of the Galactic centre. The Northern Arm shows high levels of emissive polarization at 12.5 m (peaking at 12% in the region near and to the west of IRS1) with coherent eld structures. One of the most prominent magnetic structures links the four most luminous bow-shock sources in the Northern Arm, suggesting that the out ows play an important role in sculpting the eld con guration in this region. The embedded out ow stars probably compress and push the B eld but do not overwhelm it. Other coherent structures indicate regions where gas may ow across the Northern Arm and may represent regions of in ow. In contrast, the bright regions in the East-West Bar have polarization vectors that are quite distinct from the underlying di use emission. It appears that the B eld in the lower intensity di use regions lies along the Bar, but that the bright regions near IRS2 and IRS6 are distinct regions. The intensity of the emission is only weakly correlated with the degree of polarization in the Northern Arm. This is consistent with saturated grain alignment where the degree of polarization depends on geometric e ects, including the angle of inclination of the eld to the line of sight and superposition of laments with di erent eld directions, rather than the alignment eciency. References References Aitken D.K., Roche P.F., Bailey J.A., Briggs G.P., Hough J.H., Thomas J.A.,1986, MNRAS, 218, 363 Aitken D. K., Gezari D., C. H., McCaughrean M., Roche P. F., 1991, ApJ, 380, 8 Aitken D.K., Smith C.H., Moore T.J.T., Roche P.F., 1998, MNRAS, 299, 743 Andersson B-G, Alazarian A., Vaillancourt J.E., 2015, ARAA 55, 501 Bailey, J.A., Hough, J.H., Axon, D.J., 1984, MNRAS, 208, 661 Genzel R., Eisenhauer F., Gillessen S., 2010. RevModPhys 82, 312 Gezari, D. Y, Tresch-Fienberg, R., Fazio, G. G., Ho mann, W. E, Gatley, L, Lamb, G., Shu, P. & McCreight, C., 1985. ApJ, 299, 1007 Glasse A.C.H., Aitken D.K., Roche P.F., 2003, Astron Nachr 324, 3 Hoang, T., Lazarian, A., 2016, ApJ 831, 159 Irons W.T., Lacy J.H., Richter M.J., 2012, ApJ 755, 90 Knacke, R.F., Capps, R.W., 1977, ApJ 216, 271 Packham, C., Hough J.H., Telesco C., 2005. 'Astronomical Polarimetry', ASP Conf Ser 343, 38 Paumard, T., Maillard, J-P., Morris, M., 2004, A&A, 426, 81 Roche P.F., Aitken D.K., 1985, MNRAS, 215, 425 Roche P.F., Lopez-Rodriguez E., Telesco C., Sch odel R., Packham C., 2018, MNRAS 476, 235 Sanchez-Bermudez J., Sch odel R., Alberdi A., Muzic K., Hummel C.A., Pott J.-U. 2014, A&A 567, 21 Smith C.H., Aitken D.K., Roche P.F., 1990, MNRAS, 246, 1 Smith C.H., Wright, C.M., Aitken D.K., Roche P.F., Hough, J.H., 2000, MN- RAS, 312, 327 Tanner, A., Ghez, A., Morris, M.R., Christou, J.C., 2005. ApJ, 624, 742 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Astrophysics arXiv (Cornell University)

High Resolution Imaging of the Magnetic Field in the central parsec of the Galaxy

Download PDF
9 pages

Loading next page...
 
/lp/arxiv-cornell-university/high-resolution-imaging-of-the-magnetic-field-in-the-central-parsec-of-C5B63k9KEN

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

ISSN
0032-0633
eISSN
ARCH-3330
DOI
10.1016/j.pss.2018.07.007
Publisher site
See Article on Publisher Site

Abstract

We discuss a high resolution (FWHM 0:45 arcsec) image of the emissive polar- ization from warm dust in the minispiral in the Galactic Centre and discuss the implications for the magnetic eld in the dusty laments. The image was ob- tained at a wavelength of 12.5 m with the CanariCam multimode mid-infrared imager on the Gran Telescopio Canarias. It con rms the results obtained from previous observations but also reveals new details of the polarization structures. In particular, we identify regions of coherent magnetic eld emission at position angles of  45 to the predominantly north{south run of eld lines in the North- ern Arm which may be related to orbital motions inclined to the general ow of the Northern Arm. The luminous stars that have been identi ed as bow-shock sources in the Northern Arm do not disrupt or dilute the eld but are linked by a coherent eld structure, implying that the winds from these objects may push and compress the eld but do not overwhelm it. The magnetic eld in the the low surface brightness regions in the East-West Bar to the south of SgrA* lies along the Bar, but the brighter regions generally have di erent polariza- tion position angles, suggesting that they are distinct structures. In the region of the Northern Arm sampled here, there is only a weak correlation between the intensity of the emission and the degree of polarization. This is consistent with saturated grain alignment where the degree of polarization depends on geometric e ects, including the angle of inclination of the eld to the line of sight and superposition of laments with di erent eld directions, rather than the alignment eciency. Keywords: Galactic Centre, dust, extinction, magnetic elds, polarization Preprint submitted to Elsevier August 1, 2018 arXiv:1807.11744v1 [astro-ph.GA] 31 Jul 2018 1. Introduction The central parsec of the Milky Way is a complex region hosting a range of unusual phenomena, including a supermassive black hole of mass  4 10 M , a young stellar cluster with numerous luminous mass-losing stars together with molecular and ionized gas laments mixed with dust (e.g. Genzel, Eisenhauer & Gillessen 2010). At mid-infrared wavelengths, the most prominent structure is the mini-spiral of gas and dust that comprises the Northern Arm and East- West Bar which lie within the 1.5{5 pc circumnuclear disk of molecular material around the Galactic Centre. The mid-IR emission arises from dust grains heated to temperatures of 180{250 K (e.g. Smith, Aitken & Roche 1990; Gezari et al. 1985). The derived dust temperature is fairly uniform in the bright laments but there are local peaks close to the Northern Arm out ow sources. Polarimetric measurements have revealed coherent polarization structures in the minispiral (Aitken et al., 1986, 1991, 1998; Glasse, Aitken & Roche, 2003), which provide information on the magnetic eld in the laments. The magnetic eld appears to be quite ordered and strong in the Northern Arm, with estimated eld strength of approximately 2 mGauss through indirect arguments, but much less ordered in the East-West Bar. On the assumption that the dust grains have a net align- ment to the eld lines, determination of the polarization allows us to estimate the eld directions in the plane of the sky. Aitken et al. (1998) have argued that the degree of polarization may be saturated in the Northern Arm, in which case, varations in the percentage polarization are related to the inclination of the eld with respect to the plane on the sky. Combination of polarimetric observations with radial velocities of the ionized gas then provides information on the 3-D structure of the laments. Here we discuss new higher resolution imaging polarimetry of the southern part of the Northern Arm and brightest region of the East-West Bar of the minispiral and the detailed correspondence between the polarization structures and other constituents in the central parsec. 2. Polarimetry of the Galactic Centre The centre of the Galaxy is viewed through an 8 kpc long column through the interstellar medium. This renders it invisible at short wavelengths, but the e ects of extinction are much reduced at infrared wavelengths. It appears that the extinction over the central parsec is relatively uniform and characterised by a fairly constant extinction with an optical depth at the peak of the silicate ab- sorption feature,   3:6 (e.g. Roche & Aitken 1985). Spectropolarimetry 9:7m of the compact mid-IR sources in the minispiral has revealed that the polariza- tion signatures can be interpreted by an intrinsic emissive polarization signature overlaid by a constant interstellar polarization component (Aitken et al., 1986). The polarization image presented here was measured at 12.5 m where the emissive component is large while the interstellar absorptive component, which has a pro le similar to the silicate absorption band, has decreased substantially from its peak near 10 m. The interstellar absorptive polarization is discussed 2 Absorptive Emissive Figure 1: Measured infrared spectral dependence of polarization towards IRS1, the brightest compact source in the minispiral in the mid-IR. From the top: Intensity, Percentage Polar- ization and Position Angle. Note the change in position angle with increasing wavelength from the near- to the mid-IR re ecting the change from domination by interstellar absorptive polarization to intrinsic emissive polarization. Note also the rotation in position angle across the 10m silicate band. At short infrared wavelengths and near the peak of the silicate ab- sorption feature, interstellar polarization dominates, but in the wings of the silicate band near 8 and 12 m, the emissive component is dominant. This behaviour is modelled by an emissive component at PA = 120 deg overlaid by an interstellar absorptive component at 0 deg. The observations are made in apertures of  4 arcsec. Near-IR data are from Bailey, Hough & Axon (1984); Knacke & Capps (1977) while the 8-13 and 17-22 m spectropolarimetric data are from Smith et al. 2000. 1996ASPC...97..551R in detail by Aitken et al. (1991, 1998) who compare estimates of the interstellar polarization from mid-infrared spectro- and imaging- polarimetry and conclude that all of the results are consistent with a constant absorptive polarization of 1.8% at PA= 5  5 deg at 12.5 m arising from cool aligned grains in the ISM. The observed spectral dependence of the infrared polarization measured towards IRS1 is shown in Fig. 1, where the changes in position angle reveal the way the contributions from the emissive and absorptive components change with wavelength. The detailed decomposition of the emissive and absorptive components between 8-13 m can be found in Smith et al. (2000). Because of dissipative forces in grains rotating in a magnetic eld, non- spherical dust particles will tend to align so that the grain spin axis is parallel to the maximum moment of inertia. The grain spin axis will precess around the magnetic eld, providing a net alignment of an ensemble of dust grains, so that the long axis of a grain is preferentially aligned to be perpendicular to the magnetic B eld (see e.g. Andersson, Lazarian & Vaillancourt 2015 for a review). This results in polarized emission with the position angle normal to the B eld, so that when the emissive polarization vectors are rotated by 90 , they trace out the component of the eld directions in the plane of the sky. 3. Observations The observations were made with the Canaricam mid-IR imager mounted on the Nasmyth platform of the 10.4-m Gran Telescopio Canarias. The data were obtained in the polarimetric mode wherein a rotating cryogenic half wave plate is used in conjunction with a focal plane mask and a Wollaston prism to produce simultaneous images of the e- and o-rays in three slots, each 2 x 20 arcsec (Packham, Hough & Telesco, 2005). The telescope was stepped by 1 arcsec in Right Ascension after each exposure to provide full sampling and contiguous polarization maps of the minispiral. The Galactic Centre only rises above 2 airmass for short periods above La Palma, and the delivered image quality is seeing-limited at FWHM = 0.45 arcsec. The polarization image was produced by coadding three sets of exposures which were aligned and corrected for the waveplate eciency and for instrumental polarization induced by the o - axis re ection in the Nasmyth mirror according to the prescriptions provided on the GTC/Canaricam website . The instrumental polarization induced by the tertiary mirror has been measured at 0.6% with the position angle dependent on the instrument rotator angle. Observations of unpolarized stars taken with Canaricam (though not on the same nights as the Galactic Centre) indicate residual instrumental polarization < 0:2%. Finally, the image was corrected for interstellar polarization by taking the measured polarization of the isolated luminous supergiant star, IRS3, which has a constant polarization position angle across the silicate band, to be representative of the interstellar component only (see the discussion in Aitken et al. 1998). The measured polarization of IRS3 see http://www.gtc.iac.es/instruments/canaricam 4 in the current data is 1.7% at PA 10 deg and we have used this value to correct for interstellar polarization. On the assumption that IRS3 is intrinsically unpolarised, this procedure corrects for any residual instrumental polarization as well as the interstellar absorptive component. The plate scale of Canaricam is 0.08 arcsec pixel . The data presented here have been smoothed to 0.24 arcsec pixel to improve the signal to noise ratio per point while preserving full sampling. For more details of the data analysis and processing see Roche et al. (2018). 4. Polarization image The emissive polarization map is shown in Fig. 2. In the image, the vectors have been rotated by 90 degrees, so that they point along the magnetic eld directions. It is clear that, as found previously, (e.g. Glasse, Aitken & Roche 2003), the Northern Arm, which is a coherent structure running from the north- east of the image before becoming indistinct south of SgrA* (e.g. Irons, Lacy & Richter 2012) contains substantial regions of coherent vectors and that the bright compact sources neither dilute nor disturb the local eld con gurations. This is seen more clearly and at higher resolution in the current image, where a run of vectors is seen linking the bright objects, IRS5, IRS10, IRS1 and IRS21. These objects have been identi ed as luminous stars with substantial out ows which produce bow shocks where the stellar winds impact on the Northern Arm laments (Tanner et al., 2005; Sanchez-Bermudez et al., 2014), and produce local temperature peaks (Smith, Aitken & Roche 1990). The polarization at these positions closely follows that in the adjacent laments indicating that the out ow sources are embedded in the magnetic elds and that while the out ow energy may compress and push the B elds, it does not disrupt them. The coherent vectors linking the bright embedded out ow sources suggests that these objects have played a role in pushing the magnetic ux tube. In addition to this coherent structure linking the embedded objects and the general N{S vectors following the larger scale structure of the Northern Arm near the top of the image, two regions of coherent vectors crossing the Arm at PA  45 can be identi ed to the east and south-east of IRS1. The rst of these appears to converge or merge with the vector ow immediately south of IRS1 and continue to IRS21; the second becomes indistinct where it meets a region of N{S vectors. These B eld structures may indicate regions where gas is owing across the Northern Arm and in fact they provide a qualitative match to some of the Keplerian orbits proposed by Paumard, Maillard & Morris (2004) and may indicate regions of inward gas ow. Some grain alignment mechanisms, such as radiative torques acting on grains might be expected to lead to greater alignment and higher polarization in regions with high radiation uxes (e.g. Andersson, Lazarian & Vaillancourt 2015). To further investigate this, the percentage polarization is plotted against the pixel intensity in the region of the Northern Arm to the north and east of IRS21 in Fig. 3. Regions below the threshold for reliable polarization estimation (1 Jy arcsec ) are excluded. The intensity varies by a factor of >30 in this 5 IRS5 10% IRS10 IRS3 IRS1 IRS21 IRS6 IRS2 -5 10 5 0 -5 RA Offset (arcsec) Figure 2: Polarization vectors superimposed upon a greyscale of the 12.5 m intensity image. The vectors have been rotated by 90 deg so that they point along the magnetic eld directions. Coordinates are given as o sets in arcsec from the position of SgrA* which is marked by the cross. The positions of some of the bright compact objects are indicated. The lowest contour 2 2 is at 1 Jy arcsec with further contours at 3, 6 and 9 Jy arcsec . The image size is indicated by the circle at the lower right. Hand drawn lines indicate the run of coherent polarization vectors linking the embedded bow shock sources IRS5, IRS10, IRS1 and IRS21, while the lines to the south and east of IRS1 indicate the regions that cross the Northern Arm at angles of  45 . In the East-West Bar, which runs from the south-east corner of the image to IRS6, the regions of lower intensity have polarization vectors pointing along the bar while the polarization in the brighter regions appear to arise from distinct eld con gurations. 6 15 1 10 Figure 3: Polarization plotted against intensity in the region of the Northern Arm to the north and east of IRS21. The intensity is plotted logarithmically to spread out the large number of points at the lower intensity levels. region, peaking on IRS1, while the polarization ranges up to a maximum of 14%. Such a plot has to be interpreted with caution as only the component of polarization in the plane of the sky is measured here. Furthermore, there are places where local minima appear to arise from the superposition of di erently oriented laments, and the distribution of the sources of luminosity and the 3- dimensional structure of the laments will all a ect the measured polarization. 0:23 The best t regression line is P(%)= 4 S , where S is the ux density in Jy, but the correlation coecient is very low with R = 0.10. Inspection of Fig. 3 suggests at most a modest increase in polarization with intensity suggesting that the e ect expected from radiative torque alignment is modest in the central region of the Milky Way. However, Aitken et al. (1998) proposed that the grain alignment in the Northern Arm may be saturated, in which case variations in the degree of polarization along the Northern Arm re ect changes in the orientation of the eld with respect to the line of sight rather than the degree of alignment. As several alignment mechanisms, including radiative torques (e.g. Hoang & Lazarian 2016) may be able to produce saturated alignment in grains with iron inclusions, this may not provide a robust test of the alignment mechanism. At rst sight, the B eld structure in the East-West Bar seems to be much less ordered than that in the Northern Arm. Inspection of the image shows that in the regions with lower emission intensity, the eld lies predominantly along the Bar with PA  130 , but in contrast to the Northern Arm, the brighter 7 regions in the Bar have polarization levels and position angles that diverge from the local di use emission. This is especially apparent south of IRS2, where the polarization vectors are approximately north-south, and quite distinct from the underlying di use vector eld which lies along the Bar. Similarly, near IRS6 at the western end of the Bar, there is a region where the polarization vectors have PA  45 close to the junction with the western arc. It seems that the bright regions in the Bar are not embedded in the general di use B eld but are regions with distinct properties. The region south of IRS2 shows strongly blue-shifted [Ne II] emission (Irons, Lacy & Richter, 2012) and has been considered as the southern extension of the Northern Arm where it re- emerges after curling below SgrA*; the polarization vectors appear to be be consistent with that interpretation. Similarly, the region near IRS6 is where the western arc connects to or is superimposed on the Bar, and again we might expect distinct B eld properties here. 5. Conclusions The 12.5 m polarization image presented here reveals new details of the magnetic eld in the central parsec of the Galactic centre. The Northern Arm shows high levels of emissive polarization at 12.5 m (peaking at 12% in the region near and to the west of IRS1) with coherent eld structures. One of the most prominent magnetic structures links the four most luminous bow-shock sources in the Northern Arm, suggesting that the out ows play an important role in sculpting the eld con guration in this region. The embedded out ow stars probably compress and push the B eld but do not overwhelm it. Other coherent structures indicate regions where gas may ow across the Northern Arm and may represent regions of in ow. In contrast, the bright regions in the East-West Bar have polarization vectors that are quite distinct from the underlying di use emission. It appears that the B eld in the lower intensity di use regions lies along the Bar, but that the bright regions near IRS2 and IRS6 are distinct regions. The intensity of the emission is only weakly correlated with the degree of polarization in the Northern Arm. This is consistent with saturated grain alignment where the degree of polarization depends on geometric e ects, including the angle of inclination of the eld to the line of sight and superposition of laments with di erent eld directions, rather than the alignment eciency. References References Aitken D.K., Roche P.F., Bailey J.A., Briggs G.P., Hough J.H., Thomas J.A.,1986, MNRAS, 218, 363 Aitken D. K., Gezari D., C. H., McCaughrean M., Roche P. F., 1991, ApJ, 380, 8 Aitken D.K., Smith C.H., Moore T.J.T., Roche P.F., 1998, MNRAS, 299, 743 Andersson B-G, Alazarian A., Vaillancourt J.E., 2015, ARAA 55, 501 Bailey, J.A., Hough, J.H., Axon, D.J., 1984, MNRAS, 208, 661 Genzel R., Eisenhauer F., Gillessen S., 2010. RevModPhys 82, 312 Gezari, D. Y, Tresch-Fienberg, R., Fazio, G. G., Ho mann, W. E, Gatley, L, Lamb, G., Shu, P. & McCreight, C., 1985. ApJ, 299, 1007 Glasse A.C.H., Aitken D.K., Roche P.F., 2003, Astron Nachr 324, 3 Hoang, T., Lazarian, A., 2016, ApJ 831, 159 Irons W.T., Lacy J.H., Richter M.J., 2012, ApJ 755, 90 Knacke, R.F., Capps, R.W., 1977, ApJ 216, 271 Packham, C., Hough J.H., Telesco C., 2005. 'Astronomical Polarimetry', ASP Conf Ser 343, 38 Paumard, T., Maillard, J-P., Morris, M., 2004, A&A, 426, 81 Roche P.F., Aitken D.K., 1985, MNRAS, 215, 425 Roche P.F., Lopez-Rodriguez E., Telesco C., Sch odel R., Packham C., 2018, MNRAS 476, 235 Sanchez-Bermudez J., Sch odel R., Alberdi A., Muzic K., Hummel C.A., Pott J.-U. 2014, A&A 567, 21 Smith C.H., Aitken D.K., Roche P.F., 1990, MNRAS, 246, 1 Smith C.H., Wright, C.M., Aitken D.K., Roche P.F., Hough, J.H., 2000, MN- RAS, 312, 327 Tanner, A., Ghez, A., Morris, M.R., Christou, J.C., 2005. ApJ, 624, 742

Journal

Astrophysics – arXiv (Cornell University)

Published: Jul 31, 2018

There are no references for this article.