Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://www.naic.edu/~tghosh/a2234/references/molecule_SO2.ps
Äàòà èçìåíåíèÿ: Mon Apr 2 21:14:02 2007
Äàòà èíäåêñèðîâàíèÿ: Mon Apr 11 12:59:52 2016
Êîäèðîâêà: IBM-866

Ïîèñêîâûå ñëîâà: arp 220
A&A 411, L465--L468 (2003)
DOI: 10.1051/0004í6361:20031442
c
# ESO 2003
Astronomy
&
Astrophysics
First detections of extragalactic SO 2 , NS and NO
S. MartÒÐn 1 , R. Mauersberger 1 , J. MartÒÐníPintado 2 , S. GarcÒÐaíBurillo 3 , and C. Henkel 4
1 Instituto de RadioastronomÒÐa MilimÒetrica (IRAM), Avda. Divina Pastora 7 NC, 18012 Granada, Spain
2 Departamento de AstrofisÒÐca Molecular e Infrarroja, Instituto de Estructura de la Materia, CSIC, Serrano 121,
28006 Madrid, Spain
3 Observatorio AstronÒomico Nacional (OAN), Apartado 1143, 28800 AlcalÒa de Henares, Madrid, Spain
4 MaxíPlanckíInstitut fØur Radioastronomie, Auf dem HØugel 69, 53121 Bonn, Germany
Received 8 August 2003 / Accepted 16 September 2003
Abstract. We report the first detections of SO 2 , NS and NO in an extragalactic source, the nucleus of the starburst
galaxy NGC 253. Five SO 2 transitions, three groups of hyperfine components of NO and five of NS were detected. All three
species show large abundances averaged over the inner 200 pc of NGC 253. With a relative abundance of a few 10 -7 , the emisí
sion of the NOmolecule is similar or even larger than that found in Galactic star forming regions. The derived relative molecular
abundances for each molecule have been compared with those of prototypical Galactic molecular clouds. These results seem to
confirm that large scale shocks dominate the chemistry of these molecules in the nucleus of NGC 253, ruling out a chemistry
dominated by PDRs for the bulk of the gas.
Key words. ISM: molecules -- galaxies: individual: NGC 253 -- galaxies: ISM -- galaxies: starburst -- galaxies: abundances
1. Introduction
Molecular lines from the central regions of galaxies are among
the most promising tools to explore and understand the hisí
tory of our universe (e.g. Combes et al. 1999 and references
therein). In many galaxies most of the molecular emission
stems from highly excited gas within the central few 100 pc.
However, the processes driving the excitation and the complex
chemistry of the gas are not entirely clear.
Molecular studies of nearby active galaxies such as
NGC 253, M 82 or Arp 220 suggest that C shocks, photodisí
sociating radiation, and irradiation by Xírays or cosmic rays
(e.g. Rigopoulou et al. 2002; GØusten et al. 1981; Farquhar et al.
1994) play an important role in the heating and the chemistry
of nuclear gas. The dominant mechanisms and their relative
importance are still unclear.
Sulfuríbearing molecules such as SO 2 and NS are present
in a wide variety of interstellar conditions, displaying enhanced
abundances in the hot cores of high mass star forming regions.
In hot cores, the chemistry of these molecules is determined
by grainímantle evaporation into warm gas (Charnley 1997).
Abundance ratios such as NS/CS have been suggested to be
one of the signatures of shocks in hot cores (Viti et al. 2001).
The NO molecule is a fairly abundant molecule in the di#use
ISM of the Milky Way. This species is the main precursor of
the N/O chemical network (Halfen et al. 2001).
In this letter we report the first detections of SO 2 , NS
and NO in an extragalactic source, the nucleus of NGC 253.
Send o#print requests to: S. MartÒÐn, eímail: martin@iram.es
This region is one of the most prolific sources of molecuí
lar emission outside the Milky Way (Mauersberger & Henkel
1993). The large amount of hot gas (Mauersberger et al. 2003)
and the high abundance of molecules with special chemistry,
such as SiO (GarcÒÐaíBurillo et al. 2000) indicates that the presí
ence of large scale shocks should dominate the heating and the
chemistry of the molecular gas. Also the very high abundance
of the three molecules reported in this letter and their excitation
can be explained if the chemistry of the nucleus of NGC 253 is
dominated by large scale shocks.
2. Observations and results
Observations of the spectral lines (Table 1) were carried out
with the IRAM 30 m telescope. Continuum measurements on
nearby sources made every #2 hours were used to keep a pointí
ing accuracy better than #3 ## . A wobbling secondary mirror
was used, with a symmetrical beam throw of 4 # in azimuth and
a switching frequency of 0.5 Hz.
As spectrometers we used two 256 ½ 4 MHz filterbanks for
the 2 mm transitions and two 512 ½ 1 MHz filterbanks for the
lines at 1.5 and 3 mm. The SIS receivers were tuned to SSB
with an image band rejection >10dB. Beam sizes were 21 ##
(at 3 mm), 17 ## (at 2 mm) and 10 ## or 12 ## (at 1.5 mm). Spectra
were calibrated with the standard dual load system. Intensities
are given on a mainíbeam brightness temperature scale (T MB ).
Linear baselines were removed from the spectra and the resultí
ing profiles are shown in Fig. 1.
Letter
to
the
Editor

L466 S. MartÒÐn et al.: First detections of extragalactic SO 2 , NS and NO
Fig. 1. Spectra of the SO 2 , NS and NO transitions observed towards
the nucleus of NGC 253 (# 1950 = 00 h 45 m 06.0 s , # 1950 = -25 # 33 # 45 ## ).
Velocity resolutions are 10, 15 and 20 km s -1 respectively. Ticks indií
cate the expected position of HF components of NS and NO. The TMB
scale of the yíaxis is in mK.
We detected five transitions of SO 2 in the 2 mm band, three
groups of hyperfine components of NO and five of NS in
the 2, 3 and 1.5 mm atmospheric windows. Profiles are coní
sistent with the frequencies in the spectral line catalogues of
Lovas (1992) and Pickett et al. (1998). We checked that no
other known molecular transitions, both in the signal and the
image band, significantly contaminate the observed features.
The only line partially blended is SO 2 at 135 GHz that overlaps
with 34 SO (see Fig. 1).
3. Analysis
3.1. Derived line parameters for SO 2 , NS and NO
To derive the SO 2 line parameters, we have fitted a single
Gaussian to the observed profiles. Figure 1 shows the Gaussian
fits superposed on the observed spectra, and Table 1 presents
the line parameters derived from the fit.
Fitting the NS and NO profiles is more complex as their
ground state presents #ídoubling. Thus each rotational level J
is split into two single rotational levels with opposite parity.
We use the usual notation to label, in Table 1, the lower and
upper series of single rotational levels as e and f for NS,
and # + and # - for NO. Rotational levels are further split into
Table 1. Parameters derived from Gaussian fit to SO 2 transitions and
hyperfine structure fitting for NO and NS transitions.
SO 2
# Transition # TMB dv V LSR #v 1/2 TMB
(MHz) mK km s -1 km s -1 km s -1 mK
134 004.8 8(2,6)--8(1,7) 380 (90) 242 91 3.9
135 696.0 5(1,5)--4(0,4) 1000 (200) 245 a 140 a 5.4
140 306.1 6(2,4)--6(1,5) 650 (80) 248 115 5.3
146 605.5 4(2,2)--4(1,3) 1080 (230) 241 148 6.9
151 378.6 2(2,0)--2(1,1) 800 (200) 225 158 4.8
NS b
# Transition # TMB dv V LSR #v 1/2 TMB
(MHz) J - J # mK km s -1 km s -1 km s -1 mK
115 556.2 5
2 - 3
2
f 3400 (500) 220 a 250 a 5.6
161 297.2 7
2 - 5
2 e 2500 (700) 199 224 4.3
161 697.2 7
2 - 5
2 f 3400 (600) 249 278 4.7
207436.2 9
2 - 7
2 e 1500 (350) 220 a 250 a 2.2
207 834.9 9
2 - 7
2 f 1800 (350) 220 a 250 a 2.6
NO b
# Transition # TMB dv V LSR #v 1/2 TMB
(MHz) J - J # mK km s -1 km s -1 km s -1 mK
150 176.5 3
2 - 1
2 # + 4700 (700) 244 157 13.8
250 436.8 5
2 - 3
2 # + 3100 (320) 250 150 5.4
250 796.4 5
2 - 3
2 # - 2180 (330) 215 147 6.1
a This parameter was fixed in the fit.
b Transitions and integrated intensity refer to the whole group
of HF components while TMB and frequency refer to the main
component.
hyperfine (HF) components (Gerin et al. 1992). Figure 1 indií
cates the location of the HF components, which are unresolved
due to the broad intrinsic linewidth of the emission from the
nucleus of NGC 253. Single Gaussian fits cannot be used in
this case. We fitted the HF components simultaneously with a
comb of Gaussian profiles with identical width and the relative
frequency and line intensities fixed to the spectroscopic paramí
eters of NS and NO. Table 1 shows the derived line parameters.
3.2. Column densities and rotational temperatures
In order to derive reliable physical conditions when comparí
ing line intensities measured with di#erent beam sizes (# b ),
we have to make assumptions on the overall extent (# s ) of
the nuclear emission of SO 2 , NO and NS. According to meaí
surements of CO and CN transitions with the di#erent beam
sizes of the 30 m telescope and SEST, we derive an extent of
the emitting region of #20 ## (240 pc at a distance of 2.5 Mpc;
Mauersberger et al. 2003) This is in agreement with the size
obtained by Mauersberger et al. (2003) from the high anguí
lar resolution interferometric maps of CS J = 2-1 of Peng
et al. (1996). Thus, we can obtain the source averaged brightí
ness temperatures (T B = TMB # 2
s +# 2
b
# 2 ) over 20 ## and derive source
averaged column densities in the upper levels of the observed
transitions assuming optically thin emission. In the case of NS
and NO, this was done independently for each series. Figure 2
Letter
to
the
Editor

S. MartÒÐn et al.: First detections of extragalactic SO 2 , NS and NO L467
Fig. 2. Population diagrams for the SO 2 , NS and NO molecules. For NS, the # + (squares) and # - series (triangles) are plotted and fitted
separately. For NO, only the eíseries is shown.
Table 2. Source averaged column densities, rotational temperature and
abundances.
Molecule N T rot [X]/[H 2 ] a
(cm -2 ) (K)
SO 2 7 ½ 10 13 14 (9) 4 ½ 10 -9
NS 5 ½ 10 13 8 (1) 3 ½ 10 -9
NO 5 ½ 10 15 7 (2) 3 ½ 10 -7
a The uncertainty is a factor of 2 in both directions, due to the uní
certainty of the assumed N(H 2 ) = 1.7 ½ 10 22 cm -2 (Mauersberger et al.
2003).
shows the population diagrams for each molecule where the
level population can be described by a single rotational temí
perature represented as a solid straight line.
The derived T rot as well as total column densities and esí
timated fractional abundances are summarized in Table 2. To
determine the abundance, we have assumed the H 2 column dení
sity of 1.7 ½ 10 22 cm -2 derived by Mauersberger et al. (2003)
from their 13 CO measurements. The uncertainty of a factor
of 2 in the H 2 column density (Mauersberger et al. 2003)
does not critically a#ect the conclusions of this work. In gení
eral, the fractional abundances are high with the most outí
standing case of NO with an abundance larger than 10 -7 .
For the NS molecule, parameters derived for each series are:
NNS (e) = 4.4 ½ 10 13 cm -2 , T rot (e) = 8 K and NNS ( f ) =
6.4 ½ 10 13 cm -2 , T rot ( f ) = 7.6 K. No substantial di#erence is
found between series, in agreement with the result of Gerin
et al. (1992) derived from NO in di#erent Galactic molecular
clouds.
4. Discussion
To establish the mechanism driving the chemistry in NGC 253,
we compare in Table 3 the measured fractional abundances
of SO 2 , NS and NO with those in prototypical molecular
Galactic clouds dominated by di#erent types of chemistry. We
have selected two dark clouds to illustrate the ionímolecule
chemistry associated with quiescent gas, three hot cores to
illustrate grain surface and shock chemistry associated to masí
sive protostars and one photodissociation region (PDR) to ilí
lustrate the UV dominated chemistry produced by OB stars in
the main sequence.
Table 3. Fractional abundances compared with similar studies in
molecular clouds.
Source [SO 2 ]/[H 2 ] [NS]/[H 2 ] [NO]/[H 2 ]
10 -8 10 -9 10 -8
NGC 253 0.4 3 30
Dark Clouds
L134N 0.4 a 0.2--0.6 b 20 c
TMC1 <0.1 a 0.7--1.2 b 2.7 c
Hot Cores
Orion Hot Core 9.4 d 0.4 e 30 c
Sgr B2(N) 3 10 20 f
Sgr B2(M) 40 0.03 30 f
PDRs
Orion Bar 0.01 g 0.2 g
a Ohishi et al. (1992); b McGonagle et al. (1994); c Gerin et al. (1992,
1993); d Charnley (1997); e McGonagle et al. (1997); source averí
aged Nummelin et al. (2001); g Jansen et al. (1995).
Chemistry dominated by PDRs can be ruled out for
NGC 253 since the abundances of SO 2 and NO are much larger
than those in Galactic PDRs. It is remarkable that only one
source in Table 3, the hot core in Sgr B2(N), shows larger
or similar abundances than those observed in NGC 253. This
is surprising given the expected beam dilution in NGC 253
for molecules with a high dipole moment like NS and SO 2 .
However, it is very unlikely that hot cores dominate the emisí
sion of the three molecules since the low rotational temí
peratures derived are inconsistent with the high excitation
temperature (>70 K) expected from the dense hot cores. This
suggests that the origin of the large abundance of these
molecules must be related to a hot core like chemistry, but
in regions with moderate densities in which molecules are
subthermally excited. Molecular clouds in the center of the
Galaxy show hot core like chemistry produced by the ejection
of molecules from the grains by large scale shocks (see e.g.
MartÒÐníPintado et al. 2001). Chemistry driven by large scale
C shocks has been proposed to explain the high temperatures
and large abundances of NH 3 and SiO (GarcÒÐaíBurillo et al.
2000; Mauersberger et al. 2003). We will now discuss in detail
how the abundance of the molecules reported in this paper fits
into this picture.
Letter
to
the
Editor

L468 S. MartÒÐn et al.: First detections of extragalactic SO 2 , NS and NO
4.1. SO 2
Towards NGC 253, we find a SO 2 abundance between that
found in dark clouds and hot cores. The relatively high rotaí
tional temperature of SO 2 rules out that the bulk of its emisí
sion arises from dark clouds with their typical kinetic temperí
ature of 10 K. Within the context of hot core chemistry, SO 2 is
rapidly formed from the H 2 S evaporated from grainímantles
when T < 200 K. The same type of chemistry is expected
if H 2 S is injected into the gas phase by shocks. The observaí
tions of SO 2 towards Sgr B2 by Cummins et al. (1986) support
this idea. They found that SO 2 emission arises from two comí
ponents with very di#erent rotational temperatures, the first
one associated with the Sgr B2 hot core (T rot = 310K) and a
second, subthermally excited one, with a much lower temperaí
ture of 26 K associated with the envelope. The excitation temí
perature is somewhat larger in the envelope of Sgr B2 than in
NGC 253. However, the latter is averaged over a much larger
region. We therefore conclude that SO 2 arises from the high
temperature component observed in NH 3 by Mauersberger
et al. (2003).
4.2. NO
The large NO fractional abundance averaged over an almost
200 pc region suggests that NO emission is fairly widespread
through the whole nucleus of NGC 253. If this emission
would arise from dark clouds, all the clouds in the nucleus
of NGC 253 must be of the L134N type, contrasting with
what we find in the solar vicinity where most of the clouds
are like TMC1. However, the lack of SiO emission in dark
clouds (Ziurys et al. 1989) suggests that large scale shocks
are the origin of the NO abundances observed in NGC 253.
Like in the center of our Galaxy, NO emission from NGC 253
could arise from warm moderately dense gas, such as that
of the envelope of Sgr B2 which shows a particular chemí
istry due to strong shocks and/or irradiation by hard X rays
(MartÒÐníPintado 1997, 2000). It is worth noting that models of
X ray chemistry (Lepp & Dalgarno 1996) predict an enhanceí
ment of the NO abundance (10 -6 -10 -7 ) as observed in Sgr B2
and NGC 253. Both NGC 253 and the envelope of Sgr B2 have
enhanced SiO and NO abundances suggesting a similar origin.
4.3. NS
The NS abundance in NGC 253 is higher than in most Galactic
dark cloud cores and hot cores. McGonagle & Irvine (1997)
studied the excitation of NS in several molecular clouds and
found low excitation temperatures and moderate abundances in
the hot and very dense regions as expected from the large deí
struction rate of this radical under hot core conditions (n (H 2 ) >
10 6 cm 3 , T k > 100 K). Chemical models of hot cores indicate
an enhancement of the NS abundance relative to CS in the
presence of C shocks (Viti et al. 2001). The high NS/CS raí
tio of 0.4 (with X(CS) = 4 ½ 10 -9 ; Mauersberger et al. 2003)
agrees with the prediction of these models in which a shock
passed 10-20 ½ 10 3 years after the onset of radiative heating in
the hot core. This favors large scale shocks as the most likely
explanation for the NS abundance in NGC 253. It is interesting
to note that this NS/CS ratio is an order of magnitude above the
ratio measured by Hatchell et al. (2002) towards six hot cores.
The origin of the large scale shocks is not yet clear. High
angular resolution images of nuclear SiO emission in NGC 253
suggest di#erent origins (GarcÒ ÐaíBurillo et al. 2000). In the iní
ner circumnuclear disk, shocks from a dense molecular outflow
driven by massive protostar dominates. In view of the low exí
citation temperatures derived for the new molecules, the bulk
of the emission is probably not related to this origin. The other
proposed causes, shocks in the outer Lindblad resonance as a
consequence of a barred potential or generated by mass ejecí
tion from the disk, could explain the abundances and excitation
temperatures.
Interferometric observations of SO 2 , NO and NS are
needed to measure their spatial distribution and to establish the
origin of the large abundances of these molecules in the nuclear
region of NGC 253.
Acknowledgements. J. M.íP. has been partially supported by the
Ministerio de Ciencia Y TecnologÒÐa with grant ESP2002í01627 and
AYA2002í10113E.
References
Charnley, S. B. 1997, ApJ, 481, 396
Cummins, S. E., Linke, R. A., & Thaddeus, P. 1986, ApJS, 60, 819
Combes, F., Maoli, R., & Omont, A. 1999, A&A, 345, 369
Farquhar, P. R. A., Millar, T. J., & Herbst, E. 1994, MNRAS, 269, 641
GarcÒÐaíBurillo, S., MartÒÐníPintado, J., Fuente, A., & Neri, R. 2000,
A&A, 355, 499
Gerin, M., Viala, Y., Pauzat, F., & Ellinger, Y. 1992, A&A, 266, 463
Gerin, M., Viala, Y., & Casoli, F. 1993, A&A, 268, 212
GØusten, R., Walmsley, C. M., & Pauls, T. 1981, A&A, 103, 197
Halfen, D. T., Apponi, A. J., & Ziurys, L. M. 2001, ApJ, 561, 244
Hatchell, J., & Viti, S. 2002, A&A, 381, L33
Jansen, D. J., Spaans, M., Hogerheijde, M. R., & van Dishoeck, E. F.
1995, A&A, 303, 541
Lepp, S., & Dalgarno, A. 1996, A&A, 306, L21
Lovas, F. J. 1992, J. Phys. Chem. Ref. Data, 21, 181
MartÒÐníPintado, J., de Vicente, P., Fuente, A., & Planesas, P. 1997,
ApJ, 482, L45
MartÒÐníPintado, J., de Vicente, P., RodrÒÐguezíFernÒandez, N. J., Fuente,
A., & Planesas, P. 2000, A&A, 356, L5
MartÒÐníPintado, J., Rizzo, J. R., de Vicente, P., RodrÒÐguezíFernÒandez,
N. J., & Fuente, A. 2001, ApJ, 548, L65
Mauersberger, R., & Henkel, C. 1993, Rev. Mod. Astron., 6, 69
Mauersberger, R., Henkel, C., Wei‹, A., Peck, A. B., & Hagiwara, Y.
2003, A&A, 403, 561
McGonagle, D., Irvine, W. M., & Ohishi, M. 1994, ApJ, 422, 621
McGonagle, D., & Irvine, W. M. 1997, ApJ, 477, 711
Nummelin, A., Bergman, P., Hjalmarson, Š., et al. 2001, ApJS, 128,
213
Ohishi, M., Irvine, W. M., & Kaifu, N. 1992, IAU Symp., 150, 171
Peng, R., Zhou, S., Whiteoak, J. B., Lo, K. Y., & Sutton, E. C. 1996,
ApJ, 470, 821
Pickett, H. M., Poynter, R. L., Cohen, E. A., et al. 1998,
J. Quant. Spec. Radiat. Transf., 60, 883
Rigopoulou, D., Kunze, D., Lutz, D., Genzel, R., & Moorwood,
A. F. M. 2002, A&A, 389, 374
Viti, S., Caselli, P., Hartquist, T. W., & Williams, D. A. 2001, A&A,
370, 1017
Ziurys, L. M., Friberg, P., & Irvine, W. M. 1989, ApJ, 343, 201
Letter
to
the
Editor