1: %% manuscript produces a one-column, double-spaced document:
2: %% preprint2 produces a double-column, single-spaced document:
3: \documentclass[12pt,preprint]{aastex}
4: %\documentclass[manuscript]{aastex}
5: %\documentclass[preprint2]{aastex}
6:
7: \newcommand{\vdag}{(v)^\dagger}
8: \newcommand{\myemail}{skywalker@galaxy.far.far.away}
9:
10: %% You can insert a short comment on the title page using the command below.
11:
12: \slugcomment{}
13:
14:
15: \shorttitle{Far-Infrared OH in Sagittarius B2}
16: \shortauthors{Goicoechea \& Cernicharo}
17:
18:
19: \begin{document}
20:
21:
22: \title{Far-Infrared OH fluorescent emission in Sagittarius B2\altaffilmark{1}}
23:
24:
25: \author{Javier R. Goicoechea and Jos\'e Cernicharo\altaffilmark{2}}
26: \affil{Instituto de Estructura de la Materia. Departamento F\'{\i}sica
27: Molecular, CSIC, Serrano 121, E--28006 Madrid, Spain.}
28: \email{javier@isis.iem.csic.es}
29: \email{cerni@astro.iem.csic.es}
30:
31: \altaffiltext{1}{Based on observations with ISO,
32: an ESA project with instruments funded by ESA Member States
33: (especially the PI countries: France, Germany, the Netherlands
34: and the United Kingdom) and with participation of ISAS and NASA.}
35:
36:
37:
38: \altaffiltext{2}{Visiting Scientist, Division of Physics, Mathematics and
39: Astronomy, California Institute of Technology, MS 320-47, Pasadena, CA 91125, USA}
40:
41:
42:
43:
44: \begin{abstract}
45:
46: We present Infrared Space Observatory (ISO)
47: observations of $^{16}$OH and $^{18}$OH toward
48: Sgr B2 with a spectral resolution of $\sim$35 km s$^{-1}$.
49: The OH J=5/2$\rightarrow$3/2 and J=3/2$\rightarrow$1/2 rotational lines of
50: the $^2\Pi_{1/2}$ ladder are
51: seen in emission while the cross-ladder transitions
52: (from the $^2\Pi_{3/2}$ J=3/2 to the J=1/2, 3/2 and 5/2 levels
53: of the $^2\Pi_{1/2}$ ladder),
54: and the $^2\Pi_{3/2}$ J=5/2$\leftarrow$3/2 and J=7/2$\leftarrow$5/2 lines
55: are detected in absorption.
56: The $^{18}$OH $^2\Pi_{3/2}$ J=5/2$\leftarrow$3/2 $\Lambda$--doublet at $\sim$120 $\mu$m is
57: also observed in absorption.
58: All OH $\Lambda$--doublets are resolved (except
59: the $\sim$98 $\mu$m) and show, in addition to the strong
60: absorption at the velocity of Sgr B2,
61: several velocity components associated to
62: the gas surrounding Sgr B2 and to the foreground clouds along the
63: line of sight.
64:
65: No asymmetries in the line intensities of each doublet
66: have been observed.
67: We have modeled the observations using a non-local radiative
68: transfer code and found that the OH absorption/emission must arise
69: in a shell around Sgr B2 not resolved by the ISO/LWS beam. The gas
70: density is moderate, with upper limits of 10$^4$ cm$^{-3}$
71: and $\simeq$300 K in temperature. The OH abundance is high,
72: (2--5)$\times$10$^{-6}$.
73: We argue that a widespread photon dominated region
74: explains the enhancement of OH abundance.
75:
76:
77:
78:
79:
80: \end{abstract}
81:
82:
83: \keywords{
84: infrared: ISM: lines and bands
85: --- ISM: individual (Sgr B2)
86: --- ISM: molecules
87: --- line: identification
88: --- molecular data
89: --- radiative transfer
90: }
91:
92: \section{Introduction}
93:
94: The Sagittarius B2 complex represents an interesting burst of massive star
95: formation in the Galactic Center (GC) and may be representative of other
96: active galactic nuclei.
97: Large scale continuum maps from radio to far--IR wavelengths show that Sgr B2
98: is the brightest and among the most massive clouds of the GC (Cox \& Laureijs
99: 1989, Pierce--Price et al. 2000; Scoville, Solomon \& Penzias 1975).
100: The Sgr B2 M source is the brightest far--IR condensation of the complex with
101: a diameter of $\simeq$40$^{\prime\prime}$ at 100 $\mu$m (Goldsmith et al. 1992) and
102: has also the largest gas column density (Lis \& Goldsmith 1989).
103: These studies have shown that the
104: core is embedded in an extended and clumpy cloud of warm gas.
105: However, the observed rich chemistry in the gas in front of Sgr B2 and
106: its possible heating mechanisms are far from settled. Low velocity shocks
107: have been invoked to explain the enhanced abundances of some species such
108: as NH$_2$ or NH$_3$ (Flower, Pineau des For\^{e}ts \& Walmsley 1995)
109: which are not observed in more quiescent regions.
110: On the other hand, UV radiation can have an important effect on the gas
111: in the outer layers. This radiation may be provided by evolved stars
112: and by young massive stars in the envelope of Sgr B2 itself
113: (Mart\'{\i}n--Pintado et al. 1999).
114: In addition, the emission of several
115: ions with high excitation potential (such as [OIII] and [NIII]) is
116: extended in the Sgr B2 region (Goicoechea et al. 2003, in preparation). Thus,
117: a widespread ionized component producing photodissociation regions
118: (PDRs) in the envelope is also possible.
119: In any of these scenarios or in a combination of both,
120: O--bearing species such as H$_2$O, OH, H$_3$O$^+$ and atomic oxygen
121: are decisive for the thermal balance of the dense molecular gas
122: (Neufeld et al. 1995). In particular, the hydroxyl radical, OH, has
123: been predicted to be abundant in both scenarios.
124:
125:
126: The structure of the OH levels is depicted in Figure 1.
127: The fundamental rotational $\Lambda$--doublet ($\sim$119 $\mu$m) was first
128: detected in the direction of Sgr B2 in absorption against the thermal dust
129: emission by Storey, Watson \& Townes (1981). Each line of the doublet is so
130: optically thick that absorbs completely the continuum radiation avoiding any
131: check with the models. These lines are detected over a large area
132: (9$^\prime\times$27$^\prime$) around Sgr B2 (Cernicharo et al. 1999).
133: A detailed study of the far-IR spectrum of OH has been carried
134: out mainly in the Orion shocked region (Watson et al. 1985; Viscuso et al.
135: 1985; Melnick et al. 1987, 1990; Betz \& Boreiko 1989) but little
136: is known toward GC clouds such as the gas in the line
137: of sight toward Sgr B2.
138:
139: From the theoretical point of view, Offer \& van Dishoeck (1992; hereafter
140: OfD) and Offer, van Hemert \& van Dishoeck (1994; hereafter OfHD) have made
141: detailed calculations of the collisional cross sections between OH and H$_2$
142: and of the expected intensity of the OH far--IR lines. They predicted
143: strong asymmetries in the intensity of the OH $\Lambda$--doublets due
144: to important asymmetries in the collisional rates between OH and para-H$_2$.
145:
146:
147:
148: In this letter, we present high--resolution observations of several
149: OH $\Lambda$--doublets toward Sgr B2 involving levels up to $\sim$420 K.
150: All lines inside
151: the $^2\Pi_{1/2}$ ladder are seen in emission while all the remaining lines
152: are detected in absorption. The data have been modeled using a non-local
153: radiative transfer code in order to constrain some important physical
154: parameters of the warm outer layers of Sgr B2 and of the foreground gas.
155:
156:
157:
158: \section{Observations, Data Reduction and Results}
159:
160: Most of the pure rotational lines of OH are in the far--IR wavelength
161: coverage of the Long-Wavelength Spectrometer
162: (LWS; Clegg et al. 1996) on board ISO (Kessler et al. 1996).
163: The ISO data base observations used in this Letter are
164: 32201429,
165: 32600502,
166: 46201118
167: 46201123,
168: 46701803,
169: 47600809,
170: 47608099,
171: 49401705,
172: 50400823,
173: 50601013,
174: 50600603,
175: 50700610,
176: and 84500102.
177: The LWS/Fabry--Perot (FP) spectral resolution is $\simeq$0.015 $\mu$m
178: ($\sim$35 km s$^{-1}$) and allows resolving the $\Lambda$--doubling
179: of OH and its isotopes.
180: Several OH rotational lines
181: (intra and cross--ladder) were searched toward Sgr B2 M. Data are
182: shown in Figure 1.
183: The OH J= 5/2$\leftarrow$3/2 cross-ladder doublet at $\sim$34 $\mu$m was
184: observed with the Short-Wavelength Spectrometer (SWS; de Graauw et al. 1996) with a
185: spectral resolution of 1500 (ISO observation 28702002).
186: One component was observed with the SWS/FP with a velocity resolution of
187: $\sim$10 km s$^{-1}$ (ISO observation 46001217).
188: When compared to the grating observation, the
189: mean LWS/FP continuum flux of each line deviates by at least 20\%, similar to
190: the grating flux calibration errors (Swinyard et al. 1998). The
191: systematic uncertainties are always larger than the statistical errors
192: associated with the line signal--to--noise ratios (S/N). The best
193: calibration is obtained by correcting the continuum level of the FP
194: spectrum
195: in order to coincide with the grating level.
196: A polynomial baseline was fitted to each spectra and adopted as the
197: continuum level.
198: The effective SWS/FP aperture in the $\sim$34 $\mu$m range
199: is 17$^{\prime\prime}$$\times$40$^{\prime\prime}$,
200: while the LWS aperture is circular and $\sim$80$^{\prime\prime}$ in diameter.
201: All the SWS and LWS products were processed with
202: version 10.1 of the pipeline and analyzed using the ISO spectrometers data
203: reduction package ISAP.\footnote{The ISO Spectral Analysis Package (ISAP)
204: is a joint
205: development by the LWS and SWS Instruments Teams and Data Centers.
206: Contributing institutes are CESR, IAS, IPAC,MPE, RAL, and SRON.}
207:
208:
209: Sgr B2 M has been the target of several spectroscopic observations
210: in the far--IR. Molecular features have been always found in absorption.
211: However, OH shows a different behavior. All rotational doublets connecting
212: the $^{16}$OH $^2\Pi_{3/2}$ ground--state with the
213: rotational levels of the $^2\Pi_{1/2}$ ladder ($\sim$34, 53, 79 and 119
214: $\mu$m) are observed in absorption while the $^{16}$OH $^2\Pi_{1/2}$
215: intra-ladder rotational transitions at $\sim$98 and $\sim$163 $\mu$m are
216: observed in emission.
217:
218: The $\sim$79 and $\sim$53 $\mu$m $\Lambda$--doublets have maximum absorption
219: depths at the velocity of Sgr B2 M (V$_{LSR}$$\simeq$+50 km s$^{-1}$).
220: Nevertheless, there is
221: still a contribution from more negative velocity components associated
222: with the ``spiral arms clouds'' in the line of sight and with the gas
223: surrounding the GC, which we divide in two components ($\simeq-85$
224: and 0 km s$^{-1}$). The n(H$_2$) range found in these clouds varies
225: from 10$^2$ to 10$^4$ cm$^{-3}$ (Greaves 1995).
226: The absorption depths at V$_{LSR}$$\simeq$-85 km $s^{-1}$ are
227: $\simeq$30$\%$ and $\simeq$15$\%$
228: for the $\sim$79 and $\sim$53 $\mu$m doublets respectively.
229: These components are thought
230: to arise in the molecular gas within 1 kpc of the GC (Scoville 1972).
231: Assuming that the absorption at these velocities
232: is optically thin and considering only radiative
233: excitation (i.e., low T$_{ex}$),
234: we derive $\chi(OH)\sim10^{-6}$ for
235: $N(H_2)$$\simeq$(5--10)$\times$10$^{21}$ cm$^{-2}$ (see Neufeld et al.
236: 2000 for a distribution of $N(H_2)$ in the line of sight). The
237: opacity of the OH lines at V$_{LSR}$$\simeq$0 km s$^{-1}$
238: allows also to derive $\chi(OH)\sim5\times10^{-7}$ for
239: $N(H_2)$$\simeq$2.5$\times$10$^{22}$
240: cm$^{-2}$. This component is believed to arise in the diffuse gas at
241: galactocentric radii 3--8 kpc and in the Solar vicinity
242: (Greaves \& Williams 1994).
243: These values are higher than the typical
244: OH abundances of 10$^{-7}$ found in diffuse and translucent clouds
245: (Van Dishoeck \& Black 1986).
246:
247: The $\sim$84 $\mu$m doublet and the 34.603 $\mu$m line component (in
248: absorption) and the $^2\Pi_{1/2}$ doublets at $\sim$98 and
249: $\sim$163 $\mu$m (in emission) are strictly centered at
250: the velocity of Sgr B2 M and are decisive to understand the excitation
251: scheme of the molecule.
252: The ISO data also provide upper limits to the emission/absorption
253: of the J=3/2$\rightarrow$5/2 ($\sim$96 $\mu$m) and J=5/2$\rightarrow$5/2
254: ($\sim$48 $\mu$m) cross--ladder
255: transitions
256: and to the $^2\Pi_{3/2}$ J=9/2$\rightarrow$7/2 ($\sim$65 $\mu$m) and
257: $^2\Pi_{1/2}$ J=7/2$\rightarrow$5/2 ($\sim$71 $\mu$m)
258: $\Lambda$--doublets of 1.5\% of the continuum emission.
259: Finally, we present the detection of the fundamental $^{18}$OH $^2\Pi_{3/2}$
260: J= 5/2$\leftarrow$3/2 lines at $\sim$120 $\mu$m.
261: Apart from the feature associated with Sgr B2, there is another absorption
262: at V$_{LSR}$ $\simeq$0 km s$^{-1}$. The derived $^{18}$OH
263: column density is (2.5$\pm$0.5)$\times$10$^{13}$ cm$^{-2}$, implying
264: an $^{16}$OH/$^{18}$OH abundance ratio in this component of $\sim$500,
265: similar to the terrestrial value.
266:
267:
268: \section{Modeling Sgr B2 and Discussion}
269:
270:
271: The typical OH abundance in dense quiescent clouds is
272: (0.1-1)$\times$ 10$^{-7}$ with OH/H$_2$O ratios in the range
273: 1 to 10$^{-2}$ (Bergin et al. 1995).
274: According to the chemical models, the major contribution to the
275: enhanced OH abundance
276: comes from regions where water vapor is being rapidly photodissociated
277: (Sternberg \& Dalgarno 1995) or regions in which shock waves play a role
278: (Draine et al. 1983).
279: In the case of Sgr B2, the innermost regions of the cloud
280: are completely hidden in the mid and far-infrared due to the large dust
281: opacity. Almost all the observed far--IR OH come from the
282: external layers of the Sgr B2 cloud, i.e., from the surrounding gas at
283: high T$_K$ and from the
284: cold dark clouds and diffuse gas in the line of sight.
285:
286:
287: In order to estimate the OH abundance and the physical conditions leading
288: to the observed OH line emission in the V$_{LSR}\simeq+50$ km s$^{-1}$
289: component, we have modeled the first 20 OH rotational
290: levels using the code developed by Gonz\'alez-Alfonso \& Cernicharo
291: (1993). The hyperfine structure of OH is not included in the model.
292: The collisional cross sections of OfHD have been used
293: (kindly provided in electronic form by E.F van Dishoeck).
294: We have adopted a spherical geometry for a cloud consisting of two components:
295: a uniform core with a diameter of 25$^{\prime\prime}$ (3.2$\times$10$^{18}$ cm for
296: an assumed distance to Sgr B2 of 8.5 kpc), with a dust
297: temperature of 30 K (see Goicoechea \& Cernicharo, 2001)
298: and opacity at 80 $\mu$m of 5
299: ($\tau_{\lambda}=\tau_{80}*(80/\lambda({\mu}m)$),
300: and a shell of
301: variable thickness and distance to the core. The shell was divided
302: into 14 layers. All molecular transitions in the core are thermalized to
303: the dust temperature due to the large dust opacity in the far--IR.
304: In order to check the sensitivity of the results on the physical
305: parameters,
306: $n(H_2)$, T$_K$ and $N(OH)$
307: were varied from 10$^3$ to 10$^5$ cm$^{-3}$, 40 to 600 K, and
308: 1$\times$10$^{15}$ up to 1$\times$10$^{17}$ cm$^{-2}$ respectively.
309:
310:
311: The observations put some constraints on the size and density of the
312: shell. Excitation temperatures for the cross-ladder and $^2\Pi_{3/2}$
313: transitions have to be below the dust temperature in order to reproduce
314: the observed line absorptions. In addition, the shell thickness
315: and its distance to the core cannot be large compared to the
316: core size, otherwise limb effects introduce important
317: re-emission that could cancel the absorption produced by the gas in
318: front of the core. The results for some models are reproduced in
319: Figure 2. We have found that if the shell is placed at some distance
320: from the core, then it is difficult to reproduce the observations as
321: almost all lines appear in emission within the LWS beam
322: (80$^{\prime\prime}$). Even if the shell is contiguous to the core but
323: large in thickness, lines also appear in emission.
324: Models in Fig. 2 correspond to a total size of 42$^{\prime\prime}$
325: (5.3$\times$10$^{18}$ cm) for the core+shell cloud.
326:
327: The model labelled M$_1$
328: corresponds to T$_K$=40 K, $n(H_2)$=10$^4$ cm$^{-3}$ and
329: $\chi(OH)=2\times10^{-6}$. The cross-ladder and $^2\Pi_{3/2}$
330: transitions are in absorption while the $^2\Pi_{1/2}$ intra-ladder
331: transitions are predicted in emission.
332: However, although the intensity of the absorbing lines is found
333: to fit reasonably well the observations, emission lines are
334: underestimated. If the OH abundance is increased in order
335: to reproduce the intensity of the emitting lines then,
336: absorbing lines are poorly fitted as limb effects start to
337: dominate. This effect is already seen in the J=1/2$\leftarrow$3/2
338: cross-ladder transition at $\sim$79 $\mu$m. In particular, the $^2\Pi_{1/2}$
339: J=5/2$\rightarrow$3/2 transition at $\sim$98 $\mu$m
340: is strongly underestimated by model M$_1$. If the gas temperature
341: increases, collisions start to pump the OH levels and the intensity of the
342: $\sim$98 $\mu$m line is better fitted.
343: However, a new problem arises as strong
344: asymmetries in the line intensities of each $\Lambda$-doubling line
345: do appear at high temperature (no asymmetries have been observed within the
346: S/N ratio of the data ). These asymmetries were already predicted
347: by OfD and are due to a strong parity change propensity rule introduced
348: by collisions with para--H$_2$.
349: They can only be suppressed if the H$_2$ ortho--to--para (OTP) ratio is higher or if
350: radiative pumping dominates.
351: We have adopted a OTP ratio of 3.
352: Model M$_2$ in Fig. 2 corresponds to
353: T$_K$=150 K with otherwise the same parameters
354: as M$_1$. The intensity of the $\sim$98 $\mu$m doublet
355: is considerably enhanced with respect to the low temperature
356: model. The intensity asymmetry in this doublet is not visible
357: in the data as the doublet is unresolved with the LWS/FP
358: spectrometer.
359: The effect of the IR pumping strongly reduces the asymmetries
360: that could be present without the optically thick core.
361: Finally, model M$_3$ (thick lines in Figure 2) represents
362: T$_K$=300 K, $n(H_2)$=5$\times$10$^3$ cm$^{-3}$ and
363: $\chi(OH)=3\times10^{-6}$. Again, the strong asymmetry of the $\sim$98 $\mu$m
364: doublet is not observable due to the lack of spectral
365: resolution. However, the observed total intensity for the doublet
366: agrees well with that predicted from the model. For the other lines, M$_3$
367: reproduces qualitatively well the observed OH pattern, except
368: in the line wings in the 119 and 79 $\mu$m
369: doublets. Nevertheless,
370: cold foreground gas (not included in the model) absorbs
371: (like in M$_1$) at these
372: wavelengths, reducing the emission wing effects in these
373: lines.
374:
375: From our models it is clear that the gas surrounding Sgr B2 has to be
376: warm (see Ceccarelli et al. 2002) and must
377: have a moderate H$_2$ density. The OH abundance is rather high,
378: (2--5)$\times$10$^{-6}$. Assuming that M$_3$ represents a reasonable
379: fit to the OH data,
380: we derive $N($$^{18}$$OH)$=(6$\pm$2)$\times$10$^{13}$ cm$^{-2}$ for Sgr B2.
381: Hence, the $^{16}$O/$^{18}$O isotopic ratio is 240--280, in excellent agreement
382: with the value obtained
383: by Bujarrabal, Cernicharo \& Gu\'elin (1983) from the OH 18 cm lines and
384: a factor $\simeq$2 lower than in the V$_{LSR}$ $\simeq$0 km s$^{-1}$ component.
385: Higher T$_K$ could even be possible but the actual collisional
386: rates introduce very strong intensity asymmetries in the
387: cross-ladder transitions which are not observed. If the gas temperature
388: is higher, then its density has to be lower than the value used
389: in M$_3$ (5$\times$10$^3$ cm$^{-3}$).
390: The reason is that higher temperatures produce strong emission
391: in the OH high excitation transitions
392: ($\sim$71 and $\sim$65 $\mu$m for example, see Figure 2)
393: which is not observed. For T$_K$=600 K, $n(H_2)$ has to be
394: decreased to (1-2)$\times$10$^3$ cm$^{-3}$ and $\chi(OH)$ increased to
395: (0.5-1)$\times$10$^{-5}$ to obtain a crude fit the ISO data. A temperature
396: gradient across the shell, going from T$_K$=40 K in the innermost
397: regions to T$_K$=600 K in the external layers is probably a better
398: representation to the physical structure of the edge of the Sgr B2 cloud.
399:
400: The inferred $^{16}$OH column density in the shell,
401: (1.5--2.5)$\times$10$^{16}$ cm$^{-2}$, and that of H$_2^{16}$O
402: determined also from LWS/FP data (Cernicharo et al. 1997; 2002, in preparation)
403: leads to an OH/H$_2$O=0.1--1 abundance ratio.
404: For comparison, shock models of Flower et al. predict $\simeq$$10^{-3}$,
405: while models of dense dark cores predict $\simeq$10$^{-4}$.
406: On the other hand, the maximun OH abundance in PDR models is $\simeq$10$^{-5}$
407: with OH/H$_2$O ratios close to 10.
408: The large abundance found in Sgr B2 suggests that its external shells
409: are illuminated by a strong UV field producing a PDR, although low
410: velocity shocks (v$_S\sim$30 km $s^{-1}$),
411: unresolved by the ISO data, may be also present.
412: Many species formed during the evolution of the cold gas in Sgr B2 are
413: being now reprocessed in these regions. This warm gas is poorly traced by
414: millimeter and submillimeter observations, but it represents the strongest
415: contribution to the absorption/emission features in the far--IR spectrum
416: of Sgr B2. In addition to $^{16}$OH and $^{18}$OH, we have searched for several
417: related species such as OH$^+$ or H$_2$O$^+$. No lines from these
418: species have been found.
419:
420: The future far--IR instruments
421: on board next generation telescopes such as the
422: {\it{Herschel Space Observatory}} will allow a fast mapping of many OH
423: rotational lines in galactic and
424: extragalactic sources making OH a very useful tool in
425: deriving the physical properties of
426: GC giant molecular clouds. This could help in the analysis of starburst
427: galaxies like Arp 220.
428:
429:
430:
431: \acknowledgments
432: We thank Spanish DGES and PNIE for funding
433: support under grant ESP98--1351--E and PANAYA2000-1784
434: and C.M. Walmsley, V. Bujarrabal and J. Fischer for their comments on
435: the manuscript. We also thank our referee, E.F. van Dishoeck, for her
436: useful comments and suggestions.
437: JRG acknowledges \textit{UAM} for a pre--doctoral fellowship.
438:
439:
440: \newpage
441:
442:
443: \begin{thebibliography}{}
444:
445: \bibitem[]{Ber95}
446: Bergin, E.A., Langer, W.D., Goldsimth, P.F. 1995 ApJ, 441, 222
447:
448:
449: \bibitem[]{Betz89}
450: Betz, A.L., \& Boreiko, R.T. 1989, ApJ, 346, L101
451:
452: \bibitem[]{Buj83}
453: Bujarrabal, V., Cernicharo, J., \& Guelin, M. 1983, A\&A, 128, 355
454:
455:
456: \bibitem[]{Cec02}
457: Ceccarelli, C. et al. 2002, A\&A, 383, 603
458:
459:
460: \bibitem[]{Cer97}
461: Cernicharo, J., Lim, T., Cox, P., et al. 1997, A\&A, 323, L25
462:
463: \bibitem[]{Cer98}
464: Cernicharo, J., Orlandi, M.A., Gonz\'{a}lez-Alfonso, E., \& Leeks, S.J.
465: 1999, in ``The Universe as seen by ISO'', Eds. P. Cox \& M.F. Kessler,
466: ESA-SP 427, 655
467:
468:
469: \bibitem[]{Cle96}
470: Clegg, P.E., et al. 1996, A\&A, 315, L38
471:
472:
473: \bibitem[]{CoxLau89}
474: Cox, P., \& Laureijs, R. 1989, in ``The Center of the Galaxy'', Ed.
475: Morris, M., IAU, Sym.No. 136, 121
476:
477:
478: \bibitem[]{Gra96}
479: de Graauw, T. et al. 1996, 315, L49
480:
481: \bibitem[]{Dra83}
482: Draine, B.T., Roberge, W.G., \& Dalgarno, A. 1983, ApJ, 264, 485
483:
484: \bibitem[]{Flo83}
485: Flower, D.R., Pineau des For\^{e}ts \& M.C. Walmsley, 1995, 294, 815
486:
487:
488: \bibitem[]{Goi01a}
489: Goicoechea, J.R., \& Cernicharo, J. 2001, ApJ, 554, L213
490:
491:
492: \bibitem[]{Gol92}
493: Goldsmith, P.F., Lis, D.C., Lester, D.F., \& Harvey, P.M.
494: 1992, ApJ, 389, 338
495:
496: \bibitem[]{Gon93}
497: Gonz\'alez-Alfonso E., Cernicharo J., 1993, A.\&A., 279, 506
498:
499:
500: \bibitem[]{Gre94}
501: Greaves, J.S. \& Williams, P.G. 1994, A\&A, 290, 259
502:
503:
504: \bibitem[]{Gre95}
505: Greaves, J.S. 1995, MNRAS, 273, 918
506:
507:
508: \bibitem[]{Kes96}
509: Kessler, M.F., et al. 1996, A\&A, 315, L27
510:
511:
512: \bibitem[]{LisGold89}
513: Lis, D.C, \& Goldsmith, P.F. 1989, ApJ, 337, 704
514:
515:
516: \bibitem[]{Pin99}
517: Mart\'{\i}n--Pintado, J., Gaume, R.A., Rod\'{\i}guez--Fern\'andez,
518: N., de Vicente, P., \& Wilson, T.L. 1999, ApJ, 519, 667
519:
520: \bibitem[]{Mel87}
521: Melnick, G.J., Genzel, R., \& Lugten, J.B. 1987, ApJ, 321, 530
522:
523:
524: \bibitem[]{Mel90}
525: Melnick, G.J., Stacey, G.J., Genzel, R., Lugten, J.B. \& Poglitsch, A.
526: 1990, ApJ, 348, 161
527:
528:
529: \bibitem[]{Neu95}
530: Neufeld, D.A., Lepp, S., \& Melnick, G.J. 1995,
531: ApJS, 100, 132
532:
533:
534: \bibitem[]{Neu00}
535: Neufeld, D.A., et al. 2000, ApJ, 539, L111
536:
537: \bibitem[]{OfD92}
538: Offer, A.R., \& van Dishoeck, E.F. 1992, MNRAS, 257, 377 (OfD)
539:
540: \bibitem[]{OfHD94}
541: Offer, A.R., van Hemert, M.C., \& van Dishoeck, E.F. 1994,
542: JChPh, 100, 362 (OfHD)
543:
544:
545: \bibitem[]{Pie00}
546: Pierce--Price, D., Richer, J.S., Greaves, J.S. et al. 2000, ApJ, 545, L121
547:
548:
549: \bibitem[]{Scov72}
550: Scoville, N.Z. 1972, ApJ, 175, L127
551:
552:
553: \bibitem[]{Scov75}
554: Scoville, N.Z., Solomon, P.M. \& Penzias, A.A. 1975, ApJ, 201, 352
555:
556:
557: \bibitem[]{Ste95}
558: Sternberg, A., \& Dalgarno, A. 1995, ApJS, 99, 565
559:
560: \bibitem[]{Sto81}
561: Storey, J., Watson, D.M., \& Townes, C.H. 1981, ApJ, 244, L27
562:
563:
564: \bibitem[]{Swi98}
565: Swinyard, B.M., et al. 1998, SPIE, 3354, 888
566:
567: \bibitem[]{VDB86}
568: Van Dishoeck, E.F., \& Black, J.H. 1986, ApJS, 62, 109
569:
570:
571: \bibitem[]{Vic85}
572: Vicuso, P.J., et al. 1985, ApJ, 296, 149
573:
574: \bibitem[]{Wat85}
575: Watson, D.M., Genzel, R., Townes, C.H., \& Storey, J.W.V.
576: 1985, ApJ, 298, 316
577:
578: \end{thebibliography}
579:
580:
581: \clearpage
582:
583: \begin{figure}
584: \caption
585: {
586: ISO observations of $^{16}$OH and $^{18}$OH
587: toward Sgr B2 M and rotational energy diagram of OH. The two ladders,
588: $^2\Pi_{3/2}$ and $^2\Pi_{1/2}$, produced by the spin orbit interaction
589: are shown. The $\Lambda$--doubling splitting of each rotational
590: level has been enhanced for clarity. The hyperfine structure of the
591: rotational levels is not shown. The observed rotational transitions
592: are indicated by arrows and the observed ISO line profiles are
593: shown in boxes.
594: The intensity scale
595: corresponds to the continuum normalized flux and the abscissa to the
596: wavelength in microns. When lines are centered only at Sgr B2 velocities
597: ($\simeq$+50 km s$^{-1}$), the LSR velocity scale is not shown.
598: All $\Lambda$--doublets are LWS/FP data excepting the 34.6 $\mu$m
599: doublet for which SWS grating data ($\lambda/\Delta\lambda\simeq$1500)
600: are shown. One component has been observed with the SWS/FP
601: ($\lambda/\Delta\lambda\simeq$30000) and data are shown in the inset of
602: the corresponding box. The $^{18}$OH $^2\Pi_{3/2}$ J=5/2$\leftarrow$3/2
603: doublet is shown in the upper inset.}
604: \end{figure}
605:
606:
607: \begin{figure}
608: \plotone{f1.eps}
609: \end{figure}
610:
611:
612:
613: \clearpage
614:
615: \begin{figure}
616: \caption
617: {
618: Results from selected models of Sgr B2 component (V$_{LSR}\simeq+50$ km s$^{-1}$)
619: discussed in the text. The central panels
620: correspond to the cross-ladder transitions
621: of OH. Line intensities have been normalized to the continuum dust
622: emission. The wavelength of the transitions is indicated at the
623: right of each panel. The model labelled M$_1$ corresponds to T$_k$=40 K,
624: n(H$_2$)=10$^4$ cm$^{-2}$, $\chi_{OH}$=2$\times$10$^{-6}$.
625: The model labelled M$_2$ corresponds to T$_k$=150 K,
626: n(H$_2$)=10$^4$ cm$^{-2}$, $\chi_{OH}$=2$\times$10$^{-6}$.
627: The model labelled M$_3$ corresponds to T$_k$=300 K,
628: n(H$_2$)=5$\times$10$^3$ cm$^{-2}$, $\chi_{OH}$=3$\times$10$^{-6}$.
629: The velocity resolution in the modeled line profiles is 1 km
630: s$^{-1}$.}
631: \end{figure}
632:
633:
634: \begin{figure}
635: \plotone{f2.eps}
636: \end{figure}
637:
638: \end{document}
639:
640:
641:
642: