Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.atnf.csiro.au/research/masermeeting/web_papers/511_Brogan.pdf Дата изменения: Mon Mar 26 03:47:17 2007 Дата индексирования: Sun Dec 23 10:27:47 2007 Кодировка: Поисковые слова: molecular cloud

OH (1720 MHz) Masers: Signposts of SNR/Molecular Cloud Interactions
Crystal Brogan (NRAO)
W. M. Goss (NRAO), M. Claussen (NRAO), I. Hoffman, Anita Richards (Jodrell Bank), C. Chandler (NRAO), Anne Green (U. Sydney), Jasmina Lazendic (MIT)

IAU 242: Astrophysical Masers, Alice Springs, March 14, 2007

1


The Discovery of OH (1720 MHz) SNR Masers
A Brief History:
- (1968) Goss & Robinson observe "anomolous" OH (1720 MHz) emission toward SNRs W28, W44, & GC - (1993) Frail, Goss & Slysh identify with maser emission - (1996, 1997) SNR surveys by Frail et al.; Green et al.; Yusef-Zadeh et al. (1996, 1999); Koralesky et al. (1998)

2


The Discovery of OH (1720 MHz) SNR Masers
A Brief History:
Dame et al. 2001

- (1968) Goss & Robinson observe "anomolous" OH (1720 MHz) emission toward SNRs W28, W44, & GC - (1993) Frail, Goss & Slysh identify with maser emission - (1996, 1997) SNR surveys by Frail et al.; Green et al.; Yusef-Zadeh et al. (1996, 1999); Koralesky et al. (1998) OH (1720 MHz) masers are found toward 10% of Galactic SNRs (~20) [and one in the LMC, Brogan et al 2003] All but one SNR OH maser is inside the Molecular Ring They are rather weak, < few 10s of Jy Has a lower flux density cutoff - at least for compact emission (e.g. methanol class II) No accompanying mainline OH or H2O masers
3

Green et al. (1997)


Properties of SNR OH (1720 MHz) Masers
· Collisional pump requires strict range of physical conditions (Wardle 1999; Lockett et al. 1999): ­ Temperature 50 to 125 K ­ Density 104 to 105 cm-3 · These conditions are easily met when a Ctype SNR shock hits a molecular cloud ­ X-rays from SNR help dissociate H2O · Only shocks (more or less) transverse to our line of site give enough velocity coherence · Can get magnetic field strength from the Zeeman effect (Z=0.65 Hz/µG) Provides only means of directly observing strength of B-field in SNRs (but only if we can figure out the proper conversion) Th

OH Energy Levels

Dipole Selection Rule F = 0, ±1

ese masers probe SNR/ molecular cloud interactions
4


Simplified Model of SNR/Molecular Cloud Interaction
Maser emission zone

Based on Wardle (1999) and Lockett et al. (1999) models

A few notes about directions and angles ·The shock compression suggests that the B-field should be in the plane of the shock ­ only this component is amplified · Linear P.A. · Linear P.A.
maser synch

can be || or to the magnetic field is to the magnetic field
5


SNR (1720 MHz ) Properties (< 2002)
Simple Zeeman patterns with B = 0.2 ­ 5 mG and weak (~ 10%) linear polarization Magnetic pressure ram pressure B-field appears to be stronger with higher resolution Except GC, scatter broadening not important and they are not significantly variable Significant internal Faraday depolarization unlikely because Faraday length > gain length Show excellent correlation with density/shock tracing molecular gas Follow-up revealed shocked molecular gas in previously unknown cases
Claussen et al. (1997, 1999, 2002); Koralesky et al. (1998); Frail & Mitchell (1998); Yusef-Zadeh et al. (1999); Brogan et al. (2000)
Greyscale: CO (1-0) emission

G349.7+0.2

Reynoso & Mangum (2000)

CO (3-2)

CTB37A
6


Open Questions
· What are the detailed properties of the polarization and can we distinguish between theoretical models? · How is the maser flux distributed on small size scales and what are the brightness temperatures? · Does the B-field really increase with higher resolution which might be indicative of more tangled B-fields on larger size scales? · Are these masers saturated?

7


Zeeman Effect in SNR OH Masers
SNR OH (1720) maser line splitting is not fully resolved so that:

I V Z B dI/dv

V ~ c Z B dI/2d But for these masers the line splitting/line width ~ 0.1 i.e. NOT splitting << line width this case has not been studied in detail, limiting analysis may not apply Could be different than thermal case where c=cos (Elitzur 1998; Watson & Wyld 2001)

The keys to understanding the B/maser relationship: · linear polarization · high enough S/N to measure line splitting between R-L directly
8


OH (1720 MHz) Masers in W28

Red = MSX 8 m Green = SGPS+VLA 20cm Blue = VLA 90cm
Brogan et al. (2005) Hoffman et al. (2005a) 9


Morphology of W28 Masers
MERLIN VLBA VLBA

At D~2.5 kpc 50 mas = 2 x 1015 cm (125 AU)
Merlin Vlsr range = 1.4 km/s VLBA Vlsr range = 0.1 km/s

~50% of VLA flux recovered by MERLIN data ~70-85% of MERLIN flux recovered by VLBA data
Hoffman et al. (2005a)
10


W28 Linear Polarization
CO (3-2)
P.A.maser compared to P.A.synch

P.I. = 2% ­ 20% P.A.
Frail & Mitchell (1998); Hoffman et al. (2005a)

maser

Vectors rotated by 90o
11


OH (1720 MHz) Masers in W44 CO (3-2)

P.A.

synch

Red = MIPSGAL 24 m Green = GLIMPSE 8 m Blue = VLA 90cm
Castelletti et al. (2007)

P.I. = 7% ­ 14%
P.A.
maser

vectors NOT rotated by 90
Hoffman et al. (2005b)

o

12


OH (1720 MHz) Masers in W51B
HII: W51B HII: W51A

+
SNR: W51C

90cm VLA Red = MSX 8 m Green = VLA 20cm Blue = VLA 90cm

One of most luminous SFRs in Galaxy

Located at the tangent point of Sagittarius Arm lots of material piled up along the line of sight; very complex kinematics
Brogan et al. (in prep.)
13


MERLIN & VLBA Toward W51C Maser
· VLBA Resolution 12.5 x 6.3 mas · At MERLIN and VLBA scales, both regions are missing about half of the VLA flux density
d ~ 3.5 x 1015 cm Tb ~3.1 x 109 K

B = 1.7 ± 0.1 mG B = 1.5 ± 0.2 mG

MERLIN
d ~ 1.2 x cm 10 K Tb ~ 1.6 x 10 B = 2.2 ± 0.1 mG 1015

B = 1.5 ± 0.03 mG

P.I. = 2% ­ 5%
B = 1.9 ± 0.04 mG
Beam ~225 x 125 mas

Brogan (2003); Brogan et al., in prep.

14


Spitzer mid-IR and JCMT Observations
Velocity resolution is ~1 km/s and extent 50 to 90 km/s

+ +

Red = GLIMPSE 8 m Green = GLIMPSE 4.5 Blue = GLIMPSE 3 m

CO(3-2) Integrated Intensity
m

White contours show 90cm VLA

Like W44, P.A. is || to shock front
15


W51 in Radio and X-rays
Koo et al. (2002)
ROSAT Soft X-rays 330 MHz VLA

Koo et al. (2002)

ASCA Hard X-rays

Previously undiscovered SNR in front of W51C responsible for masers
16

74 MHz VLA


Are OH (1720 MHz) SNR Masers Saturated?
Saturated Stokes I=I
th

v = Doppler width

Stokes Vth= b dIth/d

Unsaturated
Stokes I=Ithe
() ()

v << Doppler width = (b dIth/d) e
()

Stokes V= Vthe

Most likely somewhat saturated · Goodness of fit · High Brightness temperatures · Non-variability

17


An Uncomfortable Coincidence?
q = 1 - 2/(3sin2) GKK for completely saturated case For all SNR OH (1720 Mz) masers to date, q = 2 ­ 20% implying ~ 60 degrees for ALL 3 sources
W28

W44 & W51B

However, P.A.maser or || to field breaks degeneracy
Q positive for case and 10% Q negative for || case and 10% 60
o o

40

Cos()

Watson & Wylde (2001) and not inconsistent with Elitzur (1996, 1998)
18


An Interesting Correlation...
W28 W44

BRL = (R

CP

­ LCP)/Z

B = V/(c Z dI/2d)
19


· What are the detailed properties of the polarization and can we distinguish between theoretical models?
- Not quite there yet but it is encouraging

Conclusions

· How is the maser flux distributed on small size scales and what are the brightness temperatures?
- Core-Halo structure with about Ѕ of flux missing at MERLIN/VLBA scales (see J. Hewitt talk) - Tb ~ 10
9-10

K

· Does the B-field really increase with higher resolution which might be indicative of more tangled B-fields on larger size scales?
- No, this is purely a spatial/spectral blending issue, but maybe it does indicate something interesting about turbulence

+ +
20

· Are these masers saturated?
- Yes, at least moderately so