1: %%
2: %% Beginning of file 'sample.tex'
3: %%
4: %% Modified 2005 December 5
5: %%
6: %% This is a sample manuscript marked up using the
7: %% AASTeX v5.x LaTeX 2e macros.
8:
9: %% The first piece of markup in an AASTeX v5.x document
10: %% is the \documentclass command. LaTeX will ignore
11: %% any data that comes before this command.
12:
13: %% The command below calls the preprint style
14: %% which will produce a one-column, single-spaced document.
15: %% Examples of commands for other substyles folloxw. Use
16: %% whichever is most appropriate for your purposes.
17: %%
18: %%\documentclass[12pt,preprint]{aastex}
19:
20: %% manuscript produces a one-column, double-spaced document:
21:
22: %\documentclass[manuscript]{aastex}
23: \documentclass{emulateapj}
24: \usepackage{graphicx}
25: \usepackage{natbib}
26: \usepackage{aalongtable}
27:
28: %% preprint2 produces a double-column, single-spaced document:
29:
30: %% \documentclass[preprint2]{aastex}
31:
32: %% Sometimes a paper's abstract is too long to fit on the
33: %% title page in preprint2 mode. When that is the case,
34: %% use the longabstract style option.
35:
36: %% \documentclass[preprint2,longabstract]{aastex}
37:
38: %% If you want to create your own macros, you can do so
39: %% using \newcommand. Your macros should appear before
40: %% the \begin{document} command.
41: %%
42: %% If you are submitting to a journal that translates manuscripts
43: %% into SGML, you need to follow certain guidelines when preparing
44: %% your macros. See the AASTeX v5.x Author Guide
45: %% for information.
46:
47: \newcommand{\vdag}{(v)^\dagger}
48: \newcommand{\myemail}{beuther@mpia.de}
49:
50: %% You can insert a short comment on the title page using the command below.
51:
52: \slugcomment{Astronomical Journal in press}
53:
54: %% If you wish, you may supply running head information, although
55: %% this information may be modified by the editorial offices.
56: %% The left head contains a list of authors,
57: %% usually a maximum of three (otherwise use et al.). The right
58: %% head is a modified title of up to roughly 44 characters.
59: %% Running heads will not print in the manuscript style.
60:
61: \shorttitle{Chemical Diversity in High-Mass Star Formation}
62: \shortauthors{Beuther et al.}
63:
64: %% This is the end of the preamble. Indicate the beginning of the
65: %% paper itself with \begin{document}.
66:
67: \begin{document}
68:
69: %% LaTeX will automatically break titles if they run longer than
70: %% one line. However, you may use \\ to force a line break if
71: %% you desire.
72:
73: \title{Chemical Diversity in High-Mass Star Formation}
74:
75: %% Use \author, \affil, and the \and command to format
76: %% author and affiliation information.
77: %% Note that \email has replaced the old \authoremail command
78: %% from AASTeX v4.0. You can use \email to mark an email address
79: %% anywhere in the paper, not just in the front matter.
80: %% As in the title, use \\ to force line breaks.
81:
82: \author{H.~Beuther\altaffilmark{1}, Q.~Zhang\altaffilmark{2},
83: E.A.~Bergin\altaffilmark{3}, and T.K. Sridharan\altaffilmark{2}}
84: \altaffiltext{1}{Max-Planck-Institute for Astronomy, K\"onigstuhl 17,
85: 69117 Heidelberg, Germany}
86: \altaffiltext{2}{Harvard-Smithsonian Center for Astrophysics, 60 Garden Street,
87: Cambridge, MA 02138, USA}
88: \altaffiltext{3}{University of Michigan, 825 Dennison Building,
89: 500 Church Street, Ann Arbor, MI 48109-1042}
90: %\affil{Astronomy Department,
91: % University of California, Berkeley, CA 94720}
92:
93: %\author{C. D. Biemesderfer\altaffilmark{4,5}}
94: %\affil{National Optical Astronomy Observatories, Tucson, AZ 85719}
95: %\email{aastex-help@aas.org}
96:
97: %\and
98:
99: %\author{R. J. Hanisch\altaffilmark{5}}
100: %\affil{Space Telescope Science Institute, Baltimore, MD 21218}
101:
102: %% Notice that each of these authors has alternate affiliations, which
103: %% are identified by the \altaffilmark after each name. Specify alternate
104: %% affiliation information with \altaffiltext, with one command per each
105: %% affiliation.
106:
107: %\altaffiltext{1}{Visiting Astronomer, Cerro Tololo Inter-American Observatory.
108: %CTIO is operated by AURA, Inc.\ under contract to the National Science
109: %Foundation.}
110: %\altaffiltext{2}{Society of Fellows, Harvard University.}
111: %\altaffiltext{3}{present address: Center for Astrophysics,
112: % 60 Garden Street, Cambridge, MA 02138}
113: %\altaffiltext{4}{Visiting Programmer, Space Telescope Science Institute}
114: %\altaffiltext{5}{Patron, Alonso's Bar and Grill}
115:
116: %% Mark off your abstract in the ``abstract'' environment. In the manuscript
117: %% style, abstract will output a Received/Accepted line after the
118: %% title and affiliation information. No date will appear since the author
119: %% does not have this information. The dates will be filled in by the
120: %% editorial office after submission.
121:
122: \begin{abstract}
123: Massive star formation exhibits an extremely rich chemistry.
124: However, not much evolutionary details are known yet, especially at
125: high spatial resolution. Therefore, we synthesize previously
126: published Submillimeter Array high-spatial-resolution spectral line
127: observations toward four regions of high-mass star formation that
128: are in various evolutionary stages with a range of luminosities.
129: Estimating column densities and comparing the spatially resolved
130: molecular emission allows us to characterize the chemical evolution
131: in more detail. Furthermore, we model the chemical evolution of
132: massive warm molecular cores to be directly compared with the data.
133: The four regions reveal many different characteristics. While some
134: of them, e.g., the detection rate of CH$_3$OH, can be explained by
135: variations of the average gas temperatures, other features are
136: attributed to chemical effects. For example, C$^{34}$S is observed
137: mainly at the core-edges and not toward their centers because of
138: temperature-selective desorption and successive gas-phase chemistry
139: reactions. Most nitrogen-bearing molecules are only found toward the
140: hot molecular cores and not the earlier evolutionary stages,
141: indicating that the formation and excitation of such complex
142: nitrogen-bearing molecules needs significant heating and time to be
143: fully developed. Furthermore, we discuss the observational
144: difficulties to study massive accretion disks in the young deeply
145: embedded phase of massive star formation. The general potential and
146: limitations of such kind of dataset are discussed, and future
147: directions are outlined. The analysis and modeling of this source
148: sample reveals many interesting features toward a chemical
149: evolutionary sequence. However, it is only an early step, and many
150: observational and theoretical challenges in that field lie ahead.
151: \end{abstract}
152:
153: %% Keywords should appear after the \end{abstract} command. The uncommented
154: %% example has been keyed in ApJ style. See the instructions to authors
155: %% for the journal to which you are submitting your paper to determine
156: %% what keyword punctuation is appropriate.
157:
158: \keywords{ stars: formation -- stars: early-type -- stars:
159: individual (Orion-KL, G29.96, IRAS\,23151+5912, IRAS\,05358+3543)
160: -- ISM: molecules -- ISM: lines and bands -- ISM: evolution}
161:
162: %% From the front matter, we move on to the body of the paper.
163: %% In the first two sections, notice the use of the natbib \citep
164: %% and \citet commands to identify citations. The citations are
165: %% tied to the reference list via symbolic KEYs. The KEY corresponds
166: %% to the KEY in the \bibitem in the reference list below. We have
167: %% chosen the first three characters of the first author's name plus
168: %% the last two numeral of the year of publication as our KEY for
169: %% each reference.
170:
171:
172: %% Authors who wish to have the most important objects in their paper
173: %% linked in the electronic edition to a data center may do so by tagging
174: %% their objects with \objectname{} or \object{}. Each macro takes the
175: %% object name as its required argument. The optional, square-bracket
176: %% argument should be used in cases where the data center identification
177: %% differs from what is to be printed in the paper. The text appearing
178: %% in curly braces is what will appear in print in the published paper.
179: %% If the object name is recognized by the data centers, it will be linked
180: %% in the electronic edition to the object data available at the data centers
181: %%
182: %% Note that for sources with brackets in their names, e.g. [WEG2004] 14h-090,
183: %% the brackets must be escaped with backslashes when used in the first
184: %% square-bracket argument, for instance, \object[\[WEG2004\] 14h-090]{90}).
185: %% Otherwise, LaTeX will issue an error.
186:
187:
188: \section{Introduction}
189: \label{intro}
190:
191: One of the main interests in high-mass star formation research today
192: is a thorough characterization of the expected massive accretion disks
193: (e.g., \citealt{yorke2002,krumholz2006b}). Observations of this type
194: require high spatial resolution, and disks have been suggested in
195: numerous systems, although not always with the same tracers (for a
196: recent compilation see \citealt{cesaroni2006}). A major complication
197: is the large degree of chemical diversity in these regions. This is
198: seen on large scales in terms of the large number of molecular
199: detections (e.g., \citealt{schilke1997b,vandishoeck1998}) but in
200: particular on the small scale (of the order $10^4$\,AU and even
201: smaller) where significant chemical differentiation is found (e.g.,
202: \citealt{beuther2005a}). At present, we are only beginning to
203: decipher the chemical structure of these objects at high spatial
204: resolution, whereas our understanding of the physical properties
205: better allows us to place regions into an evolutionary context (e.g.,
206: \citealt{beuther2006b,zinnecker2007}). It is not yet clear how our
207: emerging picture of the physical evolution is related to the observed
208: chemical diversity and evolution.
209:
210: Various theory groups work on the chemical evolution during massive
211: star formation (e.g.,
212: \citealt{caselli1993,millar1997,charnley1997,viti2004,nomura2004,wakelam2005b,doty2002,doty2006}),
213: and the results are promising. However, the observational database to
214: test these models against is still relatively poor. Some single-dish
215: low-spatial-resolution line surveys toward several sources do exist,
216: but they are all conducted with different spatial resolution and
217: covering different frequency bands (e.g.,
218: \citealt{blake1987,macdonald1996,schilke1997b,hatchell1998b,mccutcheon2000,vandertak2000,vandertak2003,johnstone2003,bisschop2007}).
219: Furthermore, the chemical structure in massive star-forming regions is
220: far from uniform, and at high resolution one observes spatial
221: variations between many species, prominent examples are Orion-KL,
222: W3OH/H$_2$O or Cepheus A (see, e.g.,
223: \citealt{wright1996,wyrowski1999,beuther2005a,brogan2007}).
224:
225: Single-dish studies targeted larger source samples at low spatial
226: resolution and described the averaged chemical properties of the
227: target regions (e.g., \citealt{hatchell1998b,bisschop2007}). However,
228: no consistent chemical investigation of a sample of massive
229: star-forming regions exists at high spatial resolution. To obtain an
230: observational census of the chemical evolution at high spatial
231: resolution and to build up a database for chemical evolutionary models
232: of massive star formation, it is important to establish a rather
233: uniformly selected sample of massive star-forming regions in various
234: evolutionary stages. Furthermore, this sample should be observed in
235: the same spectral setup at high spatial resolution. While the former
236: is necessary for a reliable comparison, the latter is crucial to
237: disentangle the chemical spatial variations in the complex massive
238: star-forming regions. Because submm interferometric imaging is a time
239: consuming task, it is impossible to observe a large sample in a short
240: time. Hence, it is useful to employ synergy effects and observe
241: various sources over a few years in the same spectral lines.
242:
243: We have undertaken such a chemical survey of massive molecular cores
244: containing high-mass protostars in different evolutionary stages using
245: the Submillimeter Array (SMA\footnote{The Submillimeter Array is a
246: joint project between the Smithsonian Astrophysical Observatory and
247: the Academia Sinica Institute of Astronomy and Astrophysics, and is
248: funded by the Smithsonian Institution and the Academia Sinica.},
249: \citealt{ho2004}) since 2003 in exactly the same spectral setup. The
250: four massive star-forming regions span a range of evolutionary stages
251: and luminosities: (1) the prototypical hot molecular core (HMC)
252: Orion-KL (\citealt{beuther2004g,beuther2005a}), (2) an HMC at larger
253: distance G29.96 \citep{beuther2007d}, and two regions in a presumably
254: earlier evolutionary phase, namely (3) the younger but similar
255: luminous High-Mass Protostellar Object (HMPO) IRAS\,23151+5912
256: \citep{beuther2007f} and (4) the less luminous HMPO IRAS\,05358+3543
257: \citep{leurini2007}. Although the latter two regions also have central
258: temperatures $>$100\,K, qualifying them as ``hot'', their molecular
259: line emission is considerably weaker than from regions which are
260: usually termed HMCs. Therefore, we refer to them from now on as
261: early-HMPOs (see also the evolutionary sequence in
262: \citealt{beuther2006b}). Table \ref{source_parameters} lists the main
263: physical parameters of the selected target regions.
264:
265: The SMA offers high spatial resolution (of the order $1''$) and a
266: large enough instantaneous bandwidth of 4\,GHz to sample numerous
267: molecular transitions simultaneously (e.g., $^{28}$SiO and its rarer
268: isotopologue $^{30}$SiO, a large series of CH$_3$OH lines in the
269: $v_t=0,1,2$ states, CH$_3$CN, HCOOCH$_3$, SO, SO$_2$, and many more
270: lines in the given setup, see \S\ref{data}). Each of the these
271: observations have been published separately where we provide detailed
272: discussions of the particularities of each source
273: \citep{beuther2004g,beuther2005a,beuther2007f,leurini2007}). These
274: objects span an evolutionary range where the molecular gas and ice
275: coated grains in close proximity to the forming star are subject
276: to increasing degrees of heating. In this fashion, volatiles will be
277: released from ices near the most evolved (luminous) sources altering
278: the surrounding gas-phase chemical equilibrium and molecular emission.
279: In this paper, we synthesize these data and effectively re-observe
280: these systems at identical physical resolution in order to identify
281: coherent trends. Our main goals are to search for (a) trends in the
282: chemistry as a function of evolutionary state and (b) to explore, in a
283: unbiased manner, the capability of molecular emission to trace
284: coherent velocity structures that have, in the past, been attribute to
285: Keplerian disks. We will demonstrate that the ability of various
286: tracers to probe the innermost region, where a disk should reside,
287: changes as a function of evolution.
288:
289: \begin{table*}[htb]
290: %\begin{center}
291: \caption{Source parameters}
292: \begin{tabular}{lrrrr}
293: \hline
294: \hline
295: & Orion-KL$^b$& G29.96$^b$ & 23151$^b$& 05358$^b$ \\
296: \hline
297: $L^a$\,[L$_{\odot}$] & $10^5$ & $9\times 10^4$ & $10^5$ & $10^{3.8}$ \\
298: $d$\,[pc] & 450 & 6000 & 5700 & 1800 \\
299: $M_{\rm{gas}}^{a,c}$\,[M$_{\odot}$]& 140$^d$ & 2500$^e$ & 600 & 300 \\
300: $T_{\rm{rot}}^f$\,[K] & 300 & 340 & 150 & 220 \\
301: $N_{\rm{peak}}(\rm{H_2})^g$\,[cm${^-2}$] & $9\times 10^{24}$ & $6\times 10^{24}$ & $2\times 10^{24}$ & $2\times 10^{24}$ \\
302: Type & HMC & HMC & early-HMPO& early-HMPO \\
303: \hline
304: \hline
305: \end{tabular}
306: \footnotesize{~\\
307: $^a$ Luminosities and masses are derived from single-dish data. Since most regions split up into multiple sources, individual values for sub-members are lower.\\
308: $^b$ The SMA data are first published in \citealt{beuther2005a,beuther2007d,beuther2007f} and \citet{leurini2007}. Other parameters are taken from \citet{menten1995,olmi2003,sridha,beuther2002a}.\\
309: $^c$ The integrated masses should be accurate within a factor 5 \citep{beuther2002a}.\\
310: $^d$ This value was calculated from the 870\,$\mu$m flux of \citet{schilke1997b} following \citet{hildebrand1983} assuming an average temperature of 50\,K.\\
311: $^e$ This value was calculated from the 850\,$\mu$m flux of \citet{thompson2006} following \citet{hildebrand1983} as in comment $b$.\\
312: $^f$ Peak rotational temperatures derived from CH$_3$OH.\\
313: $^g$ H$_2$ column densities toward the peak positions derived from the submm dust continuum observatios \citep{beuther2004g,beuther2007d,beuther2007f,beuther2007c}.}
314: \label{source_parameters}
315: %\end{center}
316: \end{table*}
317:
318: \section{Data}
319: \label{data}
320:
321: The four sources were observed with the SMA between 2003 and 2005 in
322: several array configurations achieving (sub)arcsecond spatial
323: resolution. For detailed observational description, see
324: \citet{beuther2005a,beuther2007d,beuther2007f} and
325: \citet{leurini2007}. The main point of interest to be mentioned here
326: is that all four regions were observed in exactly the same spectral
327: setup. The receivers operated in a double-sideband mode with an IF
328: band of 4-6\,GHz so that the upper and lower sideband were separated
329: by 10\,GHz. The central frequencies of the upper and lower sideband
330: were 348.2 and 338.2\,GHz, respectively. The correlator had a
331: bandwidth of 2\,GHz and the channel spacing was 0.8125\,MHz, resulting
332: in a nominal spectral resolution of $\sim$0.7\,km\,s$^{-1}$. However,
333: for the analysis presented below, we smoothed the data-cubes to
334: 2\,km\,s$^{-1}$. The spatial resolution of the several datasets is
335: given in Table \ref{resolution}. Line identifications were done in an
336: iterative way: we first compared our data with the single-dish line
337: survey of Orion by \citet{schilke1997b} and then refined the analysis
338: via the molecular spectroscopy catalogs of JPL and the Cologne
339: database of molecular spectroscopy CDMS
340: \citep{poynter1985,mueller2002}.
341:
342: \begin{table}[htb]
343: \caption{Spatial resolution}
344: \begin{tabular}{lrrrr}
345: \hline
346: \hline
347: & Orion-KL & G29.96 & 23151 & 05358 \\
348: \hline
349: Cont. [$''$] & $0.8\times 0.7$ & $0.4\times 0.3$ & $0.6\times 0.5$ & $1.1\times 0.6$ \\
350: Av. Cont. [AU] & 340 & 2100 & 3100 & 1500 \\
351: Line [$''$] & $1.4\times 1.1 $& $0.6\times 0.5$ & $1.1\times 0.8$ & $1.4\times 0.8$\\
352: Av. Line [AU] & 560 & 3300 & 5400 & 2000 \\
353: \hline
354: \hline
355: \end{tabular}
356: \label{resolution}
357: \end{table}
358:
359: \section{Results and Discussion}
360:
361: \subsection{Submm continuum emission}
362:
363: To set the four regions spatially into context, Figure
364: \ref{cont_sample} presents the four submm continuum images obtained
365: with the SMA. As one expects in a clustered mode of massive star
366: formation, all regions exhibit multiple structures with the number of
367: sub-sources $\geq 2$. As discussed in the corresponding papers, while
368: most of the submm continuum peaks likely correspond to a
369: embedded protostellar sources, this is not necessarily the case for
370: all of them. Some of the submm continuum peaks could be
371: externally heated gas clumps (e.g., the Orion hot core peak,
372: \citealt{beuther2004g}) or may be produced by shock interactions with
373: molecular outflows (e.g., IRAS\,05358+3543, \citealt{beuther2007c}).
374: In the following spectral line images, we will always show the
375: corresponding submm continuum map in grey-scale as a reference frame.
376:
377: \begin{figure*}[htb]
378: \includegraphics[angle=-90,width=\textwidth]{f1.eps}
379: \caption{862\,$\mu$m continuum images toward the four massive
380: star-forming regions. Positive and negative features are shown as
381: solid and dashed contours, respectively. The contouring always
382: starts at the $3\sigma$ levels and continues in $2\sigma$ steps
383: (with $1\sigma$ values of 35, 21, 5, 7\,mJy\,beam$^{-1}$,
384: respectively). These figures are adaptions from the papers by
385: \citet{beuther2005a,beuther2007d,beuther2007f,beuther2007c}.
386: Scale-bars are presented in each panel, and the wedges show the flux
387: density scales in Jy. The axis are labeled in R.A. (J2000) and Dec.
388: (J2000). The synthesized beams sizes are listed in Table
389: \ref{resolution}.}
390: \label{cont_sample}
391: \end{figure*}
392:
393: \subsection{Spectral characteristics as a function of evolution}
394:
395: Because the four regions are at different distances (Table
396: \ref{source_parameters}), to compare the overall spectra we smoothed
397: all datasets to the same linear spatial resolution of $\sim$5700\,AU.
398: Figure \ref{sample_spectra} presents the final spectra extracted at
399: this common resolution toward the peak positions of all four regions.
400: Furthermore, we present images at the original spatial resolution of
401: the four molecular species or vibrationally-torsionally excited lines
402: that are detected toward all four target regions (Figs.~3 to 6).
403:
404: \begin{figure*}[htb]
405: \includegraphics[angle=-90,width=0.5\textwidth]{f2a.eps}
406: \includegraphics[angle=-90,width=0.5\textwidth]{f2b.eps}\\
407: \includegraphics[angle=-90,width=0.5\textwidth]{f2c.eps}
408: \includegraphics[angle=-90,width=0.5\textwidth]{f2d.eps}\\
409: \includegraphics[angle=-90,width=0.5\textwidth]{f2e.eps}
410: \includegraphics[angle=-90,width=0.5\textwidth]{f2f.eps}\\
411: \includegraphics[angle=-90,width=0.5\textwidth]{f2g.eps}
412: \includegraphics[angle=-90,width=0.5\textwidth]{f2h.eps}
413: \caption{SMA spectra extracted from the final data-cubes in the image
414: domain toward four massive star-forming regions (Orion-KL top-row,
415: G29.96 second row, IRAS\,23151+5912 third row and IRAS\,05358+3543
416: fourth row). For a better comparison, the data-cubes were all
417: smoothed to the same spatial resolution of $\sim$5700\,AU from the
418: current IRAS\,23151+3543 dataset. The circles in
419: Fig.~\ref{ch3oh_sample} outline the corresponding spatial regions.
420: The spectral resolution in all spectra is 2\,km/s.}
421: \label{sample_spectra}
422: \end{figure*}
423:
424: The spectral characteristics between the HMCs and the early-HMPOs vary
425: considerably. Table \ref{linelistall} lists all detected lines in the
426: four regions with their upper energy levels $E_u/k$ and peak
427: intensities $S_{\rm{peak}}$ at a spatial resolution of $\sim$5700\,AU.
428:
429: \subsubsection{Excitation and optical depth effects?}
430: \label{excitation_effects}
431:
432: Since the line detections and intensities are not only affected by the
433: chemistry but also by excitation effects, we have to estimate
434: quantitatively how much the latter can influence our data. In local
435: thermodynamic equilibrium, the line intensities are to first order
436: depending on the Boltzmann factor $e^{\frac{-E_u}{kT}}$ and the
437: partition function $Q$:
438:
439: $$\int I(T) dv \propto \frac{e^{\frac{-E_u}{kT}}}{Q(t)}$$
440:
441: where $\frac{E_u}{k}$ is the upper level energy state. For polyatomic
442: molecules in the high-temperature limit, one can approximate
443: $Q(T)\propto \sqrt{T^3}$ (e.g., \citealt{blake1987}). To get a feeling
444: how much temperature changes affect lines with different
445: $\frac{E_u}{k}$, one can form the ratio of $\int I(T) dv$ at two
446: different temperatures:
447:
448: $$\frac{\int I(2T) dv}{\int I(T) dv} = \frac{\sqrt{e^{\frac{E_u}{kT}}}}{\sqrt{2^3}}$$
449:
450: Equating this ratio now for a few respective upper level energy states
451: and gas temperatures of $T=100$ \& 50\,K (Table \ref{excite}), we find
452: that the induced intensity changes mostly barely exceed a factor 2.
453: Only for very highly excited lines like
454: CH$_3$OH$(7_{1,7}-6_{1,6})v_t=1$ at relatively low temperatures
455: ($T=50$\,K) do excitation effects become significant. However, at such
456: low temperatures, these highly excited lines emit well below our
457: detection limits, hence this case is not important for the present
458: comparison. Therefore, excitation plays only a minor role in
459: producing the molecular line differences discussed below, and other
460: effects like the chemistry turn out to be far more important.
461:
462: \begin{table}[htb]
463: \caption{Excitation effects}
464: \begin{tabular}{lrrr}
465: \hline
466: \hline
467: Line & $\frac{E_u}{k}$ & $\frac{\int I(2T) dv}{\int I(T) dv}$ & $\frac{\int I(2T) dv}{\int I(T) dv}$\\
468: & (K) & @ $T$=100K & @ $T$=50K\\
469: \hline
470: C$^{34}$S(7--6) & 65 & 0.49 & 0.68 \\
471: SO$_2(18_{4,1}-18_{3,1})$ & 197 & 0.95 & 2.5 \\
472: CH$_3$OH$(7_{1,7}-6_{1,6})v_t=1$& 356 & 2.1 & 12.4 \\
473: ~\\
474: \hline
475: \hline
476: \end{tabular}
477: \label{excite}
478: \end{table}
479:
480: As shown in Table \ref{linelistall}, our spectral setup barely
481: contains few lines from rarer isotopologues. We therefore cannot
482: readily determine the optical depth of the molecular lines. While it
483: is likely that, for example, the ground state CH$_3$OH lines have
484: significant optical depths, rarer species and vibrationally excited
485: lines should be more optically thin. We checked this for a few
486: respective species (e.g., C$^{34}$S or SO$_2$) via running
487: large-velocity gradient models (LVG, \citealt{vandertak2007}) with
488: typical parameters for these kind of regions, confirming the overall
489: validity of optically thin emission for most lines (see
490: \S\ref{column_para} \& \S\ref{c34s}). However, without additional
491: data, we cannot address this issue in more detail.
492:
493: \subsubsection{Column densities}
494: \label{column_para}
495:
496: Estimating reliable molecular column densities and/or abundances is a
497: relatively difficult task for interferometric datasets like those
498: presented here. The data are suffering from missing short spacings and
499: filter out large fractions of the gas and dust emission. Because of
500: the different nature of the sources and their varying distances, the
501: spatial filtering affects each dataset in a different way.
502: Furthermore, because of spatial variations between the molecular gas
503: distributions and the dust emission representing the H$_2$ column
504: densities, the spatial filtering affects the dust continuum and the
505: spectral line emission differently. On top of this, Figures
506: \ref{ch3oh_sample} to \ref{so2_sample} show that is several cases the
507: molecular line and dust continuum emission are even spatially offset,
508: preventing the estimation of reliable abundances.
509:
510: While these problems make direct abundance estimates relative to H$_2$
511: impossible, nevertheless, we are at least able to estimate approximate
512: molecular column densities for the sources. Since we are dealing with
513: high-density regions, we can derive the column densities from the
514: spectra shown in Figure \ref{sample_spectra} assuming local
515: thermodynamic equilibrium (LTE) and optically thin emission. We
516: modeled the molecular emission of each species separately using the
517: XCLASS superset to the CLASS software developed by Peter Schilke
518: (priv.~comm.). This software package uses the line catalogs from JPL
519: and CDMS \citep{poynter1985,mueller2001}. The main free parameters for
520: the molecular spectra are temperature, source size and column density.
521: We used the temperatures given in Table \ref{source_parameters}
522: (except of Orion-KL where we used 200\,K because of the large
523: smoothing to 5700\,AU) with approximate source sizes estimated from
524: the dust continuum and spectral line maps. Then we produced model
525: spectra with the column density as the remaining free parameter.
526: Considering the missing flux and the uncertainties for temperatures
527: and source sizes, the derived column densities should be taken with
528: caution and only be considered as order-of-magnitude estimates. Table
529: \ref{column} presents the results for all sources and detected
530: molecules.
531:
532: \subsubsection{General differences}
533:
534: An obvious difference between the four regions is the large line
535: forest observed toward the two HMCs Orion-KL and G29.96 and the
536: progressively less detected molecular lines toward IRAS\,23151+5912
537: and IRAS\,05358+3543. Especially prominent is the difference in
538: vibrationally-torsionally excited CH$_3$OH lines: we detect many
539: transitions in the HMCs and only a single one in the two early-HMPOs.
540: Since the vibrationally-torsionally excited CH$_3$OH lines have higher
541: excitation levels $E_u/k$, this can be relatively easily explained by
542: on average lower temperatures of the molecular gas in the early-HMPOs.
543: Assuming an evolutionary sequence, we anticipate that the two
544: early-HMPOs will eventually develop similar line forests like the two
545: HMCs.
546:
547: Analyzing the spatial distribution of CH$_3$OH we find that it is
548: associated with several physical entities (Figs.~\ref{ch3oh_sample} \&
549: \ref{ch3oh_vt1_sample}). While it shows strong emission toward most
550: submm continuum peaks, it exhibits additional interesting features.
551: For example, the double-peaked structure in G29.96
552: (Fig~\ref{ch3oh_sample}) may be caused by high optical depth of the
553: molecular emission, whereas the lower optical depth
554: vibrationally-torsionally excited lines do peak toward the central
555: dust and gas core (Fig~\ref{ch3oh_vt1_sample}, smoothing the submm
556: continuum map to the lower spatial resolution of the line data, the
557: four submm sources merge into one central peak). In contrast, toward
558: Orion-KL the strongest CH$_3$OH features in the ground state and the
559: vibrationally-torsionally excited states are toward the south-western
560: region called the compact ridge. This is the interface between a
561: molecular outflow and the ambient gas. Our data confirm previous work
562: which suggests an abundance enrichment (e.g., \citealt{blake1987}).
563: For instance in quiescent gas in Orion the ratio of CH$_3$OH/C$^{34}$S
564: $<$\,20 \citep{bergin1997}, while our data find a ratio of 100 (see
565: Table \ref{column}). This is believed to be caused by outflow shock
566: processes in the dense surrounding gas (e.g., \citealt{wright1996}).
567: Furthermore, as will be discussed in \S\ref{disks}, there exist
568: observational indications in two of the sources (IRAS\,23151+5912 and
569: IRAS\,05358+3543) that the vibrationally-torsionally excited CH$_3$OH
570: may be a suitable tracer of inner rotating disk or tori structures.
571:
572: \begin{figure*}[htb]
573: \includegraphics[angle=-90,width=\textwidth]{f3.eps}
574: \caption{CH$_3$OH contour images (line blend between
575: CH$_3$OH$(7_{2,5}-6_{2,4})$ and CH$_3$OH$(7_{2,6}-6_{2,5})$) toward
576: the four massive star-forming regions. Positive and negative
577: features are shown as solid and dashed contours. The contouring is
578: done from 15 to 95\% (step 10\%) of the peak emission, and the peak
579: emission values are 7.5, 0.9, 1.0 and 0.6\,Jy\,beam$^{-1}$ from left
580: to right, respectively. The integration regimes for the four sources
581: are [5,15], [90,104], [-60,-52] and [-20,-12]\,km\,s$^{-1}$. The
582: grey-scale with dotted contours shows the submm continuum emission
583: from Fig.~\ref{cont_sample}. This figure is an adaption from the
584: papers by
585: \citet{beuther2005a,beuther2007d,beuther2007f,leurini2007}. The axis
586: are labeled in R.A. (J2000) and Dec. (J2000). The spatial
587: resolution is listed in Table \ref{resolution}. The circles
588: represent the regions of diameter 5700\,AU used for the comparison
589: spectra in Fig.~\ref{sample_spectra}.}
590: \label{ch3oh_sample}
591: \end{figure*}
592:
593: \begin{figure*}[htb]
594: \includegraphics[angle=-90,width=\textwidth]{f4.eps}
595: \caption{Rotationally-torsionally excited
596: CH$_3$OH$(7_{1,7}-6_{1,6})(v_t=1)$ images toward the four massive
597: star-forming regions. Positive and negative features are shown as
598: solid and dashed contours. The contouring is done from 15 to 95\%
599: (step 10\%) of the peak emission, and the peak emission values are
600: 7.5, 0.9, 1.0 and 0.6\,Jy\,beam$^{-1}$ from left to right,
601: respectively. The integration regimes for the four sources are
602: [3,13], [91,105], [-58,-54] and [-18,-10]\,km\,s$^{-1}$. The
603: grey-scale with dotted contours shows the submm continuum emission
604: from Fig.~\ref{cont_sample}. This figure is an adaption from the
605: papers by
606: \citet{beuther2005a,beuther2007d,beuther2007f,leurini2007}. The axis
607: are labeled in R.A. (J2000) and Dec. (J2000). The spatial resolution
608: is listed in Table \ref{resolution}.}
609: \label{ch3oh_vt1_sample}
610: \end{figure*}
611:
612:
613: Toward G29.96, we detect most lines previously also observed toward
614: Orion-KL, a few exceptions are the $^{30}$SiO line, some of the
615: vibrationally-torsionally state $v_t=2$ CH$_3$OH lines, some N-bearing
616: molecular lines from larger molecules like CH$_3$CH$_2$CN or
617: CH$_3$CHCN, as well as a few $^{34}$SO$_2$ and HCOOCH$_3$ lines.
618: While for some of the weaker lines this difference may partly be
619: attributed to the larger distance of G29.96, for other lines such an
620: argument is unlikely to hold. For example, the CH$_3$CH$_2$CN line at
621: 348.55\,GHz is stronger than the neighboring H$_2$CS line in Orion-KL
622: whereas it remains undetected in G29.96 compared to the strong H$_2$CS
623: line there. This is also reflected in the different abundance ratio of
624: CH$_3$CH$_2$CN/H$_2$CS which is more than an order of magnitude larger
625: in Orion-KL compared with G29.96 (see Table \ref{column}). Therefore,
626: these differences are likely tracing true chemical variations between
627: sources. In contrast, the only lines observed toward G29.96 but not
628: detected toward Orion-KL are a few CH$_3$OCH$_3$ lines.
629:
630: The main SiO isotopologue $^{28}$SiO(8--7) is detected in all sources
631: but IRAS\,05358+3543. This is relatively surprising because SiO(2--1)
632: is strong in this region \citep{beuther2002d}, and the upper level
633: energy of the $J=8-7$ of $\sim$75\,K does not seem that
634: extraordinarily high to produce a non-detection. For example, the
635: detected CH$_3$OH$(7_{1,7}-6_{1,6})v_t=1$ transition has an upper
636: level energy of 356\,K. This implies that IRAS\,05358+3543 does have
637: warm molecular gas close to the central sources. However, the
638: outflow-components traced by SiO are at on average lower temperatures
639: (probably of the order 30\,K, e.g., \citealt{cabrit1990}) which may be
640: the cause of the non-detection in IRAS\,05358+3543. Furthermore, the
641: critical density of the SiO(8--7) line is about two orders of
642: magnitude higher than that of the (2--1) transition. Hence, the
643: density structure of the core may cause the (8--7) non-detection in
644: IRAS\,05358+3543 as well.
645:
646:
647: While the rarer $^{30}$SiO(8--7) isotopologue is detected toward
648: Orion-KL with nearly comparable strength as the main isotopologue
649: (Fig.~\ref{sample_spectra} and \citealt{beuther2005a}), we do not
650: detect it at all in any of the other sources.
651:
652: A little bit surprising, the H$_2$CS line at 338.081\,GHz is detected
653: toward Orion-KL, G29.96 as well as the lowest luminosity source
654: IRAS\,05358+3543, however, it remains undetected toward the more
655: luminous HMPO IRAS\,23151+5912. We are currently lacking a good
656: explanation for this phenomenon because H$_2$CS is predicted by most
657: chemistry networks as a parent molecule to be found early in the
658: evolutionary sequence (e.g., \citealt{nomura2004}). The sulphur and
659: nitrogen chemistries are also peculiar in this sample, and we outline
660: some examples below.
661:
662: \subsubsection{Sulphur chemistry}
663: \label{c34s}
664:
665: The rare Carbon-sulphur isotopologue C$^{34}$S is detected toward all
666: four regions (Fig.~\ref{sample_spectra}). However, as shown in
667: Fig.~\ref{c34s_sample} C$^{34}$S does not peak toward the main submm
668: continuum peaks but is offset at the edge of the core. In the cases of
669: G29.96 and IRAS\,23151+5912 the C$^{34}$S morphology appears to wrap
670: around the main submm continuum peaks. Toward IRAS\,05358+3543
671: C$^{34}$S is also weak toward the strongest submm peak (at the eastern
672: edge of the image) but shows the strongest C$^{34}$S emission features
673: offset from a secondary submm continuum source (in the middle of the
674: image). Toward Orion-KL, weak C$^{34}$S emission is detected in the
675: vicinity of the hot core peak, whereas we find strong C$^{34}$S
676: emission peaks offset from the dust continuum emission.
677:
678: \begin{figure*}[htb]
679: \includegraphics[angle=-90,width=\textwidth]{f5.eps}
680: \caption{C$^{34}$S(7--6) images toward the four massive star-forming
681: regions. Positive and negative features are shown as solid and
682: dashed contours. The contouring is done from 15 to 95\% (step 10\%)
683: of the peak emission, and the peak emission values are 7.5, 0.9, 1.0
684: and 0.6\,Jy\,beam$^{-1}$ from left to right, respectively. The
685: integration regimes for the four sources are [-2,14], [92,104],
686: [-58,-51] and [-17,-13]\,km\,s$^{-1}$. The grey-scale with dotted
687: contours shows the submm continuum emission from
688: Fig.~\ref{cont_sample}. This figure is an adaption from the papers
689: by \citet{beuther2005a,beuther2007d,beuther2007f,leurini2007}. The
690: axis are labeled in R.A. (J2000) and Dec. (J2000). The spatial
691: resolution is listed in Table \ref{resolution}.}
692: \label{c34s_sample}
693: \end{figure*}
694:
695: To check whether our optical-thin assumption from \S\ref{column_para}
696: is valid also for C$^{34}$S we ran LVG radiative transfer models
697: (RADEX, \citealt{vandertak2007}). We started with the H$_2$ column
698: densities from Table \ref{source_parameters} and assumed a typical
699: CS/H$_2$ abundance of $10^{-8}$ with a terrestrial CS/C$^{34}$S ratio
700: of 23 \citep{wannier1980}. Above the critical density of $2\times
701: 10^7$\,cm$^{-3}$, with the given broad C$^{34}$S spectral FWHM
702: (between 5 and 12\,km\,s$^{-1}$ for the four sources), the
703: C$^{34}$(7--6) emission is indeed optically thin. Hence optical depth
704: effects are not causing these large offsets. Since furthermore the
705: line intensities depend not just on the excitation but more strongly
706: on the gas column densities, excitation effects only, as quantified in
707: \S\ref{excitation_effects}, cannot cause the observational offsets as
708: well. Therefore, chemical evolution may be more important. A likely
709: scenario is based on different desorption temperatures of molecules
710: from dust grains (e.g., \citealt{viti2004}): CS and C$^{34}$S are
711: desorbed from grains at temperatures of a few 10\,K, and at such
712: temperatures, these molecules are expected to be well correlated with
713: the dust continuum emission. Warming up further, at 100\,K H$_2$O
714: desorbs and then dissociates to OH. The OH quickly reacts with the
715: sulphur forming SO and SO$_2$ which then will be centrally peaked (see
716: \S\ref{modeling}). Toward G29.96, IRAS\,23151+5912 and
717: IRAS\,05358+3543 we find that the SO$_2$ emission is centrally peaked
718: toward the main submm continuum peaks confirming the above outlined
719: chemical scenario (Fig.~\ref{so2_sample} \&
720: \S\ref{modeling}\footnote{For G29.96 the SO$_2$ peaks are actually
721: right between the better resolved submm continuum sources. This is
722: due to the lower resolution of the line data because smoothing the
723: continuum to the same spatial resolution, they peak very close to
724: the SO$_2$ emission peaks (see, e.g., Fig.~2 in
725: \citealt{beuther2007d}).}) . To further investigate the
726: C$^{34}$S/SO$_2$ differences, we produced column density ratio maps
727: between C$^{34}$S and SO$_2$ for G29.96, IRAS\,23151+5912 and
728: IRAS\,05358+3543, assuming local thermodynamic equilibrium and
729: optically thin emission (Fig.~\ref{ratio}). To account for the spatial
730: differences that SO$_2$ is observed toward the submm peak positions
731: whereas C$^{34}$S is seen more to the edges of the cores, for the
732: column density calculations we assumed the temperatures given in Table
733: \ref{source_parameters} for SO$_2$ whereas we used half that
734: temperature for C$^{34}$S. Although the absolute ratio values are
735: highly uncertain because of the different spatial filtering properties
736: of the two molecules, qualitatively as expected, the column density
737: ratio maps have the lowest values in the vicinity of the submm
738: continuum sources and show increased emission at the core edges. The
739: case is less clear for Orion-KL which shows the strongest SO$_2$
740: emission toward the south-eastern region called compact ridge. Since
741: this compact ridge is believed to be caused by the interaction of a
742: molecular outflow with the ambient dense gas (e.g., \citealt{liu2002})
743: and SO$_2$ is known to be enriched by shock interactions with
744: outflows, this shock-outflow interaction may dominate in Orion-KL
745: compared with the above discussed C$^{34}$S/SO$_2$ scenario. For more
746: details on the chemical evolution see the modeling in
747: \S\ref{modeling}.
748:
749:
750: \begin{figure*}[htb]
751: \includegraphics[angle=-90,width=\textwidth]{f6.eps}
752: \caption{SO$_2(14_{4,14}-18_{3,15})$ images toward the four massive
753: star-forming regions. Positive and negative features are shown as
754: solid and dashed contours. The contouring is done from 15 to 95\%
755: (step 10\%) of the peak emission, and the peak emission values are
756: 7.5, 0.9, 1.0 and 0.6\,Jy\,beam$^{-1}$ from left to right,
757: respectively. Only for IRAS\,05358+1732 the contouring starts at the
758: 35\% level because of a worse signal-to-noise ratio. The integration
759: regimes for the four sources are [0,20], [94,100], [-60,-50] and
760: [-17,-13]\,km\,s$^{-1}$. The grey-scale with dotted contours shows
761: the submm continuum emission from Fig.~\ref{cont_sample}. This
762: figure is an adaption from the papers by
763: \citet{beuther2005a,beuther2007d,beuther2007f,leurini2007}. The axis
764: are labeled in R.A. (J2000) and Dec. (J2000). The spatial
765: resolution is listed in Table \ref{resolution}.}
766: \label{so2_sample}
767: \end{figure*}
768:
769: \begin{figure*}[htb]
770: \includegraphics[angle=-90,width=\textwidth]{f7.eps}
771: \caption{Column density ratios between C$^{34}$S over SO$_2$ toward
772: G29.96, IRAS\,23151+5912 and IRAS\,05358+3543 are shown in
773: grey-scale. The contours present the corresponding submm continuum
774: maps as in shown the previous figures. For IRAS\,05358+3543 we zoom
775: only into the central region to better show the ratio variations.}
776: \label{ratio}
777: \end{figure*}
778:
779: \subsubsection{Nitrogen chemistry}
780: \label{n}
781:
782: It is intriguing that we do not detect any nitrogen-bearing molecule
783: toward the two younger HMPOs IRAS\,23151+5912 and IRAS\,05358+3543
784: (Fig.\ref{sample_spectra} and Table \ref{linelistall}). This already
785: indicates that the nitrogen-chemistry needs warmer gas to initiate or
786: requires more time to proceed. To get an idea about the more subtle
787: variations of the nitrogen chemistry, one may compare some specific
788: line pairs: For example, the HN$^{13}$C/CH$_3$CH$_2$CN line blend
789: (dominated by HN$^{13}$C) and the SO$_2$ line between 348.3 and
790: 348.4\,GHz are of similar strength in the HMC Orion-KL
791: (Fig.~\ref{sample_spectra}). The same is approximately true for the
792: HMC G29.96, although SO$_2$ is relatively speaking a bit weaker there.
793: The more interesting differences arise if one contrasts with the
794: younger sources. Toward the $10^5$\,L$_{\odot}$ early-HMPO
795: IRAS\,23151+5912, we only detect the SO$_2$ line and the
796: HN$^{13}$C/CH$_3$CH$_2$CN line blend remains a non-detection in this
797: source. In the lower luminosity early-HMPO IRAS\,05358+3543, both
798: lines are not detected, although another SO$_2$ line at 338.3\,GHz is
799: detected there.
800:
801: Judging from these line ratios, one can infer that SO$_2$ is
802: relatively easy to excite early-on in the evolution of high-mass
803: star-forming regions. Other sulphur-bearing molecules like H$_2$CS or
804: CS are released even earlier from the grains, but SO-type molecules
805: are formed quickly (e.g., \citealt{charnley1997,nomura2004}). In
806: contrast to this, the non-detection of spectral lines like the
807: HN$^{13}$C/CH$_3$CH$_2$CN line blend in the early-HMPOs indicates that
808: the formation and excitation of such nitrogen-bearing species takes
809: place in an evolutionary more evolved phase. This may either be due to
810: molecule-selective temperature-dependent gas-dust desorption processes
811: or chemical network reactions requiring higher temperatures.
812: Furthermore, simulations of chemical networks show that the complex
813: nitrogen chemistry simply needs more time to be activated (e.g.,
814: \citealt{charnley2001,nomura2004}). Recent modeling by
815: \citet{garrod2006} indicates that the gradual switch-on phase of hot
816: molecular cores is an important evolutionary stage to produce complex
817: molecules. In this picture, the HMCs have switched on their heating
818: sources earlier and hence had more time to form these nitrogen
819: molecules.
820:
821: Comparing just the two HMCs, we find that the CH$_3$CH$_2$CN at
822: 348.55\,GHz is strong in Orion-KL but not detected in G29.96. Since
823: G29.96 also exhibits less vibrationally-torsionally excited CH$_3$OH
824: lines, it is likely on average still at lower temperatures than
825: Orion-KL and may hence have not formed yet all the complex
826: nitrogen-bearing molecules present already in Orion-KL.
827:
828: \subsubsection{Modeling}
829: \label{modeling}
830:
831: To examine the chemical evolution of warm regions in greater detail we
832: used the chemical model of \citet{bergin1997}. This model includes
833: gas-phase chemistry and the freeze-out to and sublimation from grain
834: surfaces. The details of this model are provided in that paper with
835: the only addition being the inclusion of water ice formation on grain
836: surfaces. The binding energies we have adopted are for bare silicate
837: grains, except for water ice which is assumed to have a binding energy
838: appropriate for hydrogen bonding between frozen water molecules
839: \citep{fraser2001}. To explore the chemistry of these hot evaporative
840: regions we have run the model for 10$^6$\,yrs with starting conditions
841: at $n_{\rm H_2} = 10^6$\,cm$^{-3}$ and
842: $T_{\rm{gas}}=T_{\rm{dust}}$=20\,K. Under these conditions most
843: gaseous molecules, excluding H$_2$, will freeze onto the grain surface
844: and the ice mantle forms, dominated by H$_2$O. This timescale
845: (10$^5$\,yrs ) is quite short, but is longer than the free-fall time
846: at this density and is chosen as a representative time that gas might
847: spend at very high density. After completion we assume that a massive
848: star forms and the gas and dust temperature is raised such that the
849: ice mantle evaporates and the gas-phase chemistry readjusts. We have
850: made one further adjustment to this model. Our data suggest that HNC
851: (a representative nitrogen-bearing species) is not detected in
852: early-HMPOs and that the release of this species (or its pre-cursor)
853: from the ice occurs during more evolved and warmer stages. Laboratory
854: data on ice desorption suggest that the process is not as simple as
855: generally used in models where a given molecule evaporates at its
856: sublimation temperature \citep{collings2004}. Rather some species
857: co-desorb with water ice and the key nitrogen-bearing species, NH$_3$,
858: falls into this category (see also \citealt{viti2004}). We have
859: therefore assumed that the ammonia evaporates at the same temperature
860: as water ice. Our initial abundances are taken from
861: \citet{aikawa1996}, except we assume that 50\% of the nitrogen is
862: frozen in the form of NH$_3$ ice. This assumption is consistent with
863: two sets of observations. First, detections of NH$_3$ in ices towards
864: YSO's find abundances of $\sim$ 2-7\% relative to H$_2$O
865: \citep{bottinelli2007}. \citet{dartois2002} find a limit of $\sim$
866: 5\% relative to H$_2$O towards other YSO's. Our assumed abundance of
867: NH$_3$ ice is 5 $\times 10^{-6}$ relative to H$_2$O, assuming an ice
868: abundance of 10$^{-4}$ (as appropriate for cold sources) this provides
869: an ammonia ice abundance of $\sim$5\%, which is consistent with ice
870: observations. Second, high resolution ammonia observations often find
871: high abundances of NH$_3$ in the gas phase towards hot cores, often as
872: high as 10$^{-5}$, relative to H$_2$ (e.g., \citealt{cesaroni1994}).
873: Pure gas phase chemistry will have some difficulty making this high
874: abundance in the cold phase and we therefore assume it is made on
875: grains during cold phases.
876:
877: \begin{figure*}[htb]
878: \includegraphics[width=\textwidth]{f8.eps}
879: \caption{The left and right columns show results from our modeling
880: of the ``cold and warm models'', respectively. The two top-panels
881: show the abundance ratios between several species important for
882: our observations versus the time. The bottom panels present the
883: abundances relative to H$_2$. Time in this figure refers to time
884: after the onset of massive star formation and hence we do not show
885: any evolution during the cold starless phase.}
886: \label{model}
887: \end{figure*}
888:
889: Figure \ref{model} presents our chemical model results at the point of
890: star ``turn on'' where the gas and dust becomes warmer and ices
891: evaporate. Two different models were explored. The first labeled
892: as $T_{dust}$=70\,K (``warm model'') is a model which is insufficient
893: to evaporate water ice, while the second, ``hot model''
894: with $T_{dust}$=150\,K, evaporates the entire ice mantle. Much of
895: the chemical variations amongst the early-HMPO and HMC phases is found
896: for CS, SO, SO$_2$, and HNC (note this line is blended with
897: CH$_3$CH$_2$CN) and we focus on these species in our plots (along with
898: H$_2$O). It is also important to reiterate that due to differences in
899: spatial sampling between the different sources as well as between line
900: and continuum emission, we cannot derive accurate abundances from
901: these data, but rather can attempt to use the models to explain
902: trends. For the warm model the main result is that the imbalance
903: created in the chemistry by the release of most species, with ammonia
904: and water remaining as ices on the grains, leads to enhanced
905: production of CS. Essentially the removal of oxygen from the gas
906: (excluding the O found in CO) allows for rapid CS production from the
907: atomic Sulfur frozen on grains during the cold phase. Thus, for
908: early-HMPOs, which might not have a large fraction of gas above the
909: water sublimation point ($T\sim110$\,K), the ratios of CS to other
910: species are quite large. In the hot model, when the temperature can
911: evaporate both H$_2$O and NH$_3$ ice, ratios between the same
912: molecules are orders of magnitude lower (Fig.\,\ref{model}). There is
913: a large jump in the water vapor abundance (and NH$_3$, which is not
914: shown) between the warm and the hot model driving the chemistry into
915: new directions. CS remains in the gas, but not with as high abundance
916: as in the warm phase and is gradually eroded into SO and ultimately
917: SO$_2$. Hence, SO$_2$ appears to be a better tracer of more evolved
918: stages. In this sense, even the early-HMPOs can be considered as
919: relatively evolved, and it will be important to extend similar studies
920: to even earlier evolutionary stages. HNC also appears as a brief
921: intermediate product of the nitrogen chemistry.
922:
923: The above picture is in qualitative agreement with our observations:
924: that CS should be a good tracer in early evolutionary states even
925: prior to our observed sample, but that it is less well suited for more
926: evolved regions like those studied here. Other sulfur-bearing
927: molecules, in particular SO$_2$, appear to better trace the warmest
928: gas near the forming star. HNC, and perhaps other nitrogen-bearing
929: species, are better tracers when the gas is warm enough to evaporate a
930: significant amount of the ice mantle.
931:
932:
933: \subsection{Searching for disk signatures}
934: \label{disks}
935:
936: While the chemical evolution of massive star-forming regions is
937: interesting in itself, one also wants to use the different
938: characteristics of molecular lines as tools to trace various physical
939: processes. While molecules like CO or SiO have regularly been used to
940: investigate molecular outflows (e.g., \citealt{arce2006}), the problem
941: to identify the right molecular tracer to investigate accretion disks
942: in massive star formation is much more severe (e.g.,
943: \citealt{cesaroni2006}). Major observational obstacles arise from the
944: fact that disk-tracing molecular lines are usually often not
945: unambiguously found only in the accretion disk, but that other
946: processes can produce such line emission as well. For example,
947: molecular lines from CN and HCN are high-density tracers and were
948: believed to be good candidates to investigate disks in embedded
949: protostellar sources (e.g., \citealt{aikawa2001}). However,
950: observations revealed that both spectral lines are strongly affected
951: by the associated molecular outflow and hence difficult to use for
952: massive disk studies (CN \citealt{beuther2004e}, HCN
953: \citealt{zhang2007}).
954:
955: As presented in \citet{cesaroni2006}, various different molecules have
956: in the past been used to investigate disks/rotational signatures in
957: massive star formation (e.g., CH$_3$CN, C$^{34}$S, NH$_3$, HCOOCH$_3$,
958: C$^{17}$O, H$_2^{18}$O, see also Fig.~\ref{disk_examples}). The data
959: presented here add three other potential disk tracers (HN$^{13}$C and
960: HC$_3$N for G29.96, \citealt{beuther2007d}, and torsionally excited
961: CH$_3$OH for IRAS\,23151+5912 and IRAS\,05358+3543
962: \citealt{beuther2007f,leurini2007}, Fig.~\ref{disk_examples}). An
963: important point to note is that in most sources only one or the other
964: spectral line exclusively allows a study of rotational motions,
965: whereas other lines apparently do not trace the warm disks. For
966: example, C$^{34}$S traces the Keplerian motion in the young source
967: IRAS\,20126+4104 whereas it does not trace the central protostars at
968: all in the sources presented here (Fig.~\ref{c34s_sample}). In the
969: contrary, HN$^{13}$C shows rotational signatures in the HMC G29.96,
970: but it remains completely undetected in the younger early-HMPO sources
971: of our sample (Figs.~\ref{disk_examples} \& \ref{sample_spectra}). As
972: discussed in sections \ref{c34s} and \ref{n}, this implies that,
973: depending on the chemical evolution, molecules like C$^{34}$S should
974: be better suited for disk studies at very early evolutionary stages,
975: whereas complex nitrogen-bearing molecules are promising in more
976: evolved hot-core-type sources. While the chemical evolution is
977: important for these molecules, temperature effects have to be taken
978: into account as well. For example, the torsionally excited CH$_3$OH
979: line traces rotating motions in IRAS\,23151+5912 and IRAS\,05358+3543
980: (\citealt{beuther2007f,leurini2007}, Fig.~\ref{disk_examples}) but it
981: is weak and difficult to detect in colder and younger sources.
982: Therefore, some initial heating is required to employ highly excited
983: lines for kinematic studies. In addition to these evolutionary
984: effects, optical depth is important for many lines. For example,
985: \citet{cesaroni1997,cesaroni1999} have shown that CH$_3$CN traces the
986: rotating structure in IRAS\,20126+4104, whereas the same molecule does
987: not indicate any rotation in IRAS\,18089-1732 \citep{beuther2005c}. A
988: likely explanation for the latter is high optical depth of the
989: CH$_3$CN submm lines \citep{beuther2005c}. Again other molecules are
990: excited in the accretion disks as well as the surrounding envelope,
991: causing confusion problems to disentangle the various physical
992: components. {\it In summary, getting a chemical rich spectral line
993: census like the ones presented here shows that one can find several
994: disk-tracing molecules in different sources, but it also implies
995: that some previously assumed good tracers are not necessarily
996: universally useful.}
997:
998: \begin{figure*}[htb]
999: \includegraphics[angle=-90,width=6.5cm]{f9a.ps}
1000: \includegraphics[angle=-90,width=4.5cm]{f9b.eps}
1001: \includegraphics[angle=-90,width=5.0cm]{f9c.ps}\\
1002: \hspace*{4.0cm}\includegraphics[angle=-90,width=4.5cm]{f9d.eps}
1003: \includegraphics[angle=-90,width=5.2cm]{f9e.eps}
1004: \caption{Examples of rotation-tracing molecules: top-left: C$^{34}$S
1005: in IRAS\,20126+4104 \citep{cesaroni1999,cesaroni2005}, top-middle:
1006: HCOOCH$_3$ in IRAS\,18089-1732 \citep{beuther2005c}, top-right:
1007: H$_2^{18}$O in AFGL2591 \citep{vandertak2006}, bottom-left: CH$_3$OH
1008: $v_t=1$ in IRAS\,23151+5912 \citep{beuther2007f}, bottom-right:
1009: HN$^{13}$C in G29.96 \citep{beuther2007d}.}
1010: \label{disk_examples}
1011: \end{figure*}
1012:
1013: The advent of broad bandpass interferometers like the SMA now
1014: fortunately allows to observe many molecular lines simultaneously.
1015: This way, one often finds a suitable rotation-tracing molecule in an
1016: observational spectral setup. Nevertheless, one has to keep in mind
1017: the chemical and physical complexity in such regions, and it is likely
1018: that in many cases only combined modeling of infall, rotation and
1019: outflow will disentangle the accretion disk from the rest of the
1020: star-forming gas and dust core.
1021:
1022: \section{Conclusion and Summary}
1023:
1024: We compiled a sample of four massive star-forming regions in different
1025: evolutionary stages with varying luminosities that were observed in
1026: exactly the same spectral setup at high angular resolution with the
1027: SMA. We estimated column densities for all sources and detected
1028: species, and we compared the spatial distributions of the molecular
1029: gas. This allows us to start investigating chemical evolutionary
1030: effects also in a spatially resolved manner. Chemical modeling was
1031: conducted to explain our observations in more detail.
1032:
1033: A general result from this comparison is that many different physical
1034: and chemical processes are important to produce the complex chemical
1035: signatures we observe. While some features, e.g., the non-detection of
1036: the rich vibrationally-torsionally excited CH$_3$OH line forest toward
1037: the two early-HMPOs can be explained by on average lower temperatures of
1038: the molecular gas compared to the more evolved HMCs, other
1039: observational characteristics require chemical evolutionary sequences
1040: caused by heating including grain-surface and gas phase reactions.
1041: Even other features are then better explained by shock-induced
1042: chemical networks.
1043:
1044: The rare isotopologue C$^{34}$S is usually not detected right toward
1045: the main submm continuum peaks, but rather at the edge of the
1046: star-forming cores. This may be explained by temperature-selective
1047: gas-desorption processes and successive gas chemistry networks.
1048: Furthermore, we find some nitrogen-bearing molecular lines to be only
1049: present in the HMCs, whereas they remain undetected at earlier
1050: evolutionary stages. This indicates that the formation and excitation
1051: of many nitrogen-bearing molecules needs considerably higher
1052: temperatures and/or more time during the warm-up phase of the HMC,
1053: perhaps relating to the fact that NH$_3$ is bonded within the water
1054: ice mantle. Although the statistical database is still too poor to
1055: set tighter constraints, these observations give the direction how one
1056: can use the presence {\it and} morphology of various molecular lines
1057: to identify and study different (chemical) evolutionary sequences.
1058:
1059: Furthermore, we discussed the observational difficulty to
1060: unambiguously use one or the other spectral line as a tracer of
1061: massive accretion disks. While some early spectral line candidates are
1062: discarded for such kind of studies by now (e.g., CN), in many other
1063: sources we find different lines exhibiting rotational velocity
1064: signatures. The observational feature that in most sources apparently
1065: only one or the other spectral line exclusively traces the desired
1066: structures has likely to be attributed to a range of effects. (1)
1067: Chemical effects, where for example C$^{34}$S may work in the youngest
1068: sources whereas some nitrogen-bearing molecules like HN$^{13}$C are
1069: better in typical HMCs. (2) Confusion from multiple gas components,
1070: mainly outflows, infall from the envelope and rotation. (3) High
1071: optical depth from many molecular lines. This implies that for future
1072: statistical studies we have to select spectral setups that comprise
1073: many molecular lines from various species. This way, one has good
1074: chances to identify for each source separately the right molecular
1075: tracer, and hence still draw statistically significant conclusions.
1076:
1077: To advance in this field and to become more quantitative, different
1078: steps are necessary. First of all, we need to establish a larger
1079: database of more sources at different evolutionary stages, in
1080: particular even younger sources, as well as with varying luminosities
1081: to better characterize the differences and similarities. From an
1082: observational and technical point of view, although the presented data
1083: are state of the art multi-wavelength and high angular resolution
1084: observations, the quantitative interpretation is still hampered by the
1085: spatial filtering of the interferometer. To become more quantitative,
1086: it is therefore necessary to complement such data with the missing
1087: short spacing information. While we have high angular resolution in
1088: all datasets with a similar baseline coverage and hence similarly
1089: covered angular scales, the broad range of distances causes a
1090: different coverage of sampled linear spatial scales. Hence the
1091: missing short spacings affect each dataset in a different fashion
1092: which is currently the main limiting factor for a better quantitative
1093: interpretation of the data. Therefore, obtaining single-dish
1094: observations in the same spectral setup and then combining them with
1095: the SMA observations is a crucial step to derive more reliable column
1096: densities and from that abundances. These parameters then can be used
1097: by theorists to better model the chemical networks, explain the
1098: observations and predict other suitable molecules for, e.g., massive
1099: disk studies.
1100:
1101:
1102: \begin{acknowledgements}
1103: H.B.~acknowledges financial support by the Emmy-Noether-Program of the
1104: Deutsche Forschungsgemeinschaft (DFG, grant BE2578).
1105: \end{acknowledgements}
1106:
1107: %\bibliography{/home/beuther/tex/bibliography}
1108: %\bibliography{/Users/henrikbeuther/paper/bibliography}
1109: %\bibliographystyle{aa} % this does the style, aa.bst necessary
1110: \input{ms.bbl}
1111:
1112: \begin{longtable}[htb]{lrrcccc}
1113: \caption{Line peak intensities $S_{\rm{peak}}$ and upper state
1114: energy levels $E_u/k$ from spectra toward peak positions of the
1115: respective massive star-forming regions (Fig.~\ref{sample_spectra}).}\\
1116: \hline \hline
1117: Freq. & Line & $E_u/k$ & $S_{\rm{peak}}$ & $S_{\rm{peak}}$ & $S_{\rm{peak}}$ & $S_{\rm{peak}}$ \\
1118: (GHz) & & (K) & (Jy) & (Jy) & (Jy) & (Jy)\\
1119: & & & Orion & G29.96 & 23151 & 05358 \\
1120: \hline
1121: 337.252 & CH$_3$OH$(7_{3,5}-6_{3,4})$A($v_t$=2) & 739 & 6.0 \\
1122: 337.274 & CH$_3$OH$(7_{4,3}-6_{4,2})$A($v_t$=2) & 695 & 7.4 \\
1123: 337.279 & CH$_3$OH$(7_{2,5}-6_{2,4})$E($v_t$=2) & 727 & 5.4 \\
1124: 337.284 & CH$_3$OH$(7_{0,7}-6_{0,6})$A($v_t$=2) & 589 & 9.9 \\
1125: 337.297 & CH$_3$OH$(7_{1,7}-6_{1,6})$A($v_t$=1) & 390 & 10.6 & 1.7 \\
1126: 337.312 & CH$_3$OH$(7_{1,6}-6_{1,5})$E($v_t$=2) & 613 & 9.2 \\
1127: 337.348 & CH$_3$CH$_2$CN$(38_{3,36}-37_{3,35})$ & 328 & 14.2 & 1.5 & & \\
1128: 337.397 & C$^{34}$S(7--6) & 65 & 12.6 & 2.0 & 0.3 & 0.6 \\
1129: 337.421 & CH$_3$OCH$_3(21_{2,19}-20_{3,18})$ & 220 & 3.0 & 0.6 \\
1130: 337.446 & CH$_3$CH$_2$CN$(37_{4,33}-36_{4,32})$ & 322 & 11.2 & 0.8 \\
1131: 337.464 & CH$_3$OH$(7_{6,1}-6_{0,0})$A($v_t$=1) & 533 & 7.2 & 1.1 \\
1132: 337.474 & UL & & 4.9 & 0.6 \\
1133: 337.490 & HCOOCH$_3(27_{8,20}-26_{8,19})$E & 267 & 6.3 & 0.7 \\
1134: 337.519 & CH$_3$OH$(7_{5,2}-6_{5,2})$E($v_t$=1) & 482 & 8.1 & 1.0 \\
1135: 337.546 & CH$_3$OH$(7_{5,3}-6_{5,2})$A($v_t$=1) & 485 & 10.0$^b$ & 1.4$^b$ & & \\
1136: & CH$_3$OH$(7_{5,2}-6_{5,1})$A$^{-}$($v_t$=1) & 485 & 10.0$^b$ & 1.4$^b$ & & \\
1137: 337.582 & $^{34}$SO$(8_8-7_7)$ & 86 & 12.2 & 2.0 & 1.1 &\\
1138: 337.605 & CH$_3$OH$(7_{2,5}-6_{2,4})$E($v_t$=1) & 429 & 9.7 & 2.4 & & \\
1139: 337.611 & CH$_3$OH$(7_{6,1}-6_{6,0})$E($v_t$=1) & 657 & 6.2$^b$ & 2.0$^b$ & & \\
1140: & CH$_3$OH$(7_{3,4}-6_{3,3})$E($v_t$=1) & 388 & 6.2$^b$ & 2.0$^b$ & & \\
1141: 337.626 & CH$_3$OH$(7_{2,5}-6_{2,4})$A($v_t$=1) & 364 & 11.0 & 1.9 & & \\
1142: 337.636 & CH$_3$OH$(7_{2,6}-6_{2,5})$A$^-$($v_t$=1) & 364 & 8.2 & 2.5 & & \\
1143: 337.642 & CH$_3$OH$(7_{1,7}-6_{1,6})$E($v_t$=1) & 356 & 10.9$^b$ & 2.9$^b$ & 0.6$^b$ & 1.1$^b$\\
1144: 337.644 & CH$_3$OH$(7_{0,7}-6_{0,6})$E($v_t$=1) & 365 & 10.9$^b$ & 2.9$^b$ & 0.6$^b$ & 1.1$^b$\\
1145: 337.646 & CH$_3$OH$(7_{4,3}-6_{4,2})$E($v_t$=1) & 470 & 10.9$^b$ & 2.9$^b$ & 0.6$^b$ & 1.1$^b$ \\
1146: 337.648 & CH$_3$OH$(7_{5,3}-6_{5,2})$E($v_t$=1) & 611 & 10.9$^b$ & 2.9$^b$ & 0.6$^b$ & 1.1$^b$ \\
1147: 337.655 & CH$_3$OH$(7_{3,5}-6_{3,4})$A($v_t$=1) & 461 & 10.8$^b$ & 2.0$^b$ & & \\
1148: & CH$_3$OH$(7_{3,4}-6_{3,3})$A$^-$($v_t$=1) & 461 & 10.8$^b$ & 2.0$^b$ & & \\
1149: 337.671 & CH$_3$OH$(7_{2,6}-6_{2,5})$E($v_t$=1) & 465 & 10.2 & 2.1 & & \\
1150: 337.686 & CH$_3$OH$(7_{4,3}-6_{4,2})$A($v_t$=1) & 546 & 9.$^b$5 & 2.0$^b$ & & \\
1151: & CH$_3$OH$(7_{4,4}-6_{4,3})$A$^-$($v_t$=1) & 546 & 9.5$^b$ & 2.0$^b$ & & \\
1152: & CH$_3$OH$(7_{5,2}-6_{5,1})$E($v_t$=1) & 494 & 9.5$^b$ & 2.0$^b$ & & \\
1153: 337.708 & CH$_3$OH$(7_{1,6}-6_{1,5})$E($v_t$=1) & 489 & 7.9 & 1.8 & & \\
1154: 337.722 & CH$_3$OCH$_3(7_{4,4}-6_{3,3})$EE & 48 & & 0.9 & & \\
1155: 337.732 & CH$_3$OCH$_3(7_{4,3}-6_{3,3})$EE & 48 & & 1.4 & &\\
1156: 337.749 & CH$_3$OH$(7_{0,7}-6_{0,6})$A($v_t$=1) & 489 & 8.7 & 1.9 & & \\
1157: 337.778 & CH$_3$OCH$_3(7_{4,4}-6_{3,4})$EE & 48 & & 1.3 & &\\
1158: 337.787 & CH$_3$OCH$_3(7_{4,3}-6_{3,4})$AA & 48 & & 1.4 & &\\
1159: 337.825 & HC$_3$N$(37-36)v_7=1$ & 629 & 14.8 & 1.4 & & \\
1160: 337.838 & CH$_3$OH$(20_{6,14}-21_{5,16})$E & 676 & 5.6 & 1.1 & & \\
1161: 337.878 & CH$_3$OH$(7_{1,6}-6_{1,5})$A($v_t$=2) & 748 & 2.7 & 0.6 & & \\
1162: 337.969 & CH$_3$OH$(7_{1,6}-6_{1,5})$A($v_t$=1) & 390 & 12.0 & 2.1 & & \\
1163: 338.081 & H$_2$CS$(10_{1,10}-9_{1,9})$ & 102 & 5.8 & 2.3 & & 0.5 \\
1164: 338.125 & CH$_3$OH$(7_{0,7}-6_{0,6})$E & 78 & 6.9 & 2.8 & 1.4 & 1.9 \\
1165: 338.143 & CH$_3$CH$_2$CN$(37_{3,34}-36_{3,33})$ & 317 & 14.4 & 0.9 & & \\
1166: 338.214 & CH$_2$CHCN$(37_{1,37}-36_{1,36})$ & 312 & 4.0 \\
1167: 338.306 & SO$_2(18_{4,1}-18_{3,1})$ & 197 & x$^c$ & 0.8 & 1.2 & 0.7 \\
1168: 338.345 & CH$_3$OH$(7_{1,7}-6_{1,6})$E & 71 & 13.4 & 2.1 & 1.3 & 2.3\\
1169: 338.405 & CH$_3$OH$(7_{6,2}-6_{6,1})$E & 244 & 13.1$^b$ & 3.0$^b$ & & \\
1170: 338.409 & CH$_3$OH$(7_{0,7}-6_{0,6})$A & 65 & 13.1$^b$ & 3.0$^b$ & 1.5 & 2.4\\
1171: 338.431 & CH$_3$OH$(7_{6,1}-6_{6,0})$E & 254 & 9.5 & 1.8 & & \\
1172: 338.442 & CH$_3$OH$(7_{6,1}-6_{6,0})$A & 259 & 11.7$^b$ & 2.7$^b$ & & 0.5$^b$ \\
1173: & CH$_3$OH$(7_{6,2}-6_{6,1})$A$^-$ & 259 & 11.7$^b$ & 2.7$^b$ & & 0.5$^b$ \\
1174: 338.457 & CH$_3$OH$(7_{5,2}-6_{5,1})$E & 189 & 8.9 & 2.0 & 0.4 & 0.6 \\
1175: 338.475 & CH$_3$OH$(7_{5,3}-6_{5,2})$E & 201 & 12.6 & 2.5 & 0.5 & \\
1176: 338.486 & CH$_3$OH$(7_{5,3}-6_{5,2})$A & 203 & 9.4$^b$ & 2.3$^b$ & 0.8$^b$ & 0.8$^b$ \\
1177: & CH$_3$OH$(7_{5,2}-6_{5,1})$A$^-$ & 203 & 9.4$^b$ & 2.3$^b$ & 0.8$^b$ & 0.8$^b$ \\
1178: 338.504 & CH$_3$OH$(7_{4,4}-6_{4,3})$E & 153 & 8.5 & 2.7 & 0.7 & 0.7 \\
1179: 338.513 & CH$_3$OH$(7_{4,4}-6_{4,3})$A$^-$ & 145 & 13.7$^b$ & 2.8$^b$ & 1.4$^b$ & 1.3$^b$ \\
1180: & CH$_3$OH$(7_{4,3}-6_{4,2})$A & 145 & 13.7$^b$ & 2.8$^b$ & 1.4$^b$ & 1.3$^b$ \\
1181: & CH$_3$OH$(7_{2,6}-6_{2,5})$A$^-$ & 103 & 13.7$^b$ & 2.8$^b$ & 1.4$^b$ & 1.3$^b$ \\
1182: 338.530 & CH$_3$OH$(7_{4,3}-6_{4,2})$E & 161 & 5.8 & 2.7 & 0.7 & 0.9 \\
1183: 338.541 & CH$_3$OH$(7_{3,5}-6_{3,4})$A$^+$ & 115 & 12.5$^b$ & 3.0$^b$ & 2.0$^b$ & 1.4$^b$ \\
1184: 338.543 & CH$_3$OH$(7_{3,4}-6_{3,3})$A$^-$ & 115 & 12.5$^b$ & 3.0$^b$ & 2.0$^b$ & 1.4$^b$ \\
1185: 338.560 & CH$_3$OH$(7_{3,5}-6_{3,4})$E & 128 & 15.6 & 2.5 & 0.9 & 0.6 \\
1186: 338.583 & CH$_3$OH$(7_{3,4}-6_{3,3})$E & 113 & 11.5 & 3.4 & 1.0 & 1.1 \\
1187: 338.612 & SO$_2(20_{1,19}-19_{2,18})$ & 199 & x$^c$ & 2.9 & 1.5 & 1.9 \\
1188: 338.615 & CH$_3$OH$(7_{1,6}-6_{1,5})$E & 86 & x$^d$ & 2.9$^d$ & 1.5$^d$ & 1.9$^d$\\
1189: 338.640 & CH$_3$OH$(7_{2,5}-6_{2,4})$A & 103 & 7.2 & 2.5 & 1.0 & 1.0 \\
1190: 338.722 & CH$_3$OH$(7_{2,5}-6_{2,4})$E & 87 & 10.2$^b$ & 3.2$^b$ & 2.1$^b$ & 2.7$^b$\\
1191: 338.723 & CH$_3$OH$(7_{2,6}-6_{2,5})$E & 91 & 10.2$^b$ & 3.2$^b$ & 2.1$^b$ & 2.7$^b$\\
1192: 338.760 & $^{13}$CH$_3$OH$(13_{7,7}-12_{7,6})$A & 206 & 4.0 & 1.1 & & \\
1193: 338.769 & HC$_3$N$(37-36)v_7=2$ & 525 & ? & ? & & \\
1194: 338.786 & $^{34}$SO$_2(14_{4,10}-14_{3,1})$ & 134 & 6.2 \\
1195: 338.886 & C$_2$H$_5$OH$(15_{7,8}-15_{6,19})$ & 162 & 5.3 & 0.8 & & \\
1196: 338.930 & $^{30}$SiO(8--7) & 73 & 24.4 \\
1197: 339.058 & C$_2$H$_5$OH$(14_{7,7}-14_{6,8})$ & 150 & x$^e$ & 0.6 & & \\
1198: 347.232 & CH$_2$CHCN$(38_{1,38}-37_{1,37})$ & 329 & 4.9 & 0.6 & & \\
1199: 347.331 & $^{28}$SiO(8--7) & 75 & 22.1 & 0.9 & 0.7 &\\
1200: 347.438 & UL & & 7.5 & & & \\
1201: 347.446 & UL & & 3.4 & 0.8 & & \\
1202: 347.478 & HCOOCH$_3(27_{1,26}-26_{1,25})$E & 247 & 4.8 \\
1203: 347.494 & HCOOCH$_3(27_{5,22}-26_{5,21})$A & 247 & 2.9 & 0.6 & & \\
1204: 347.590 & HCOOCH$_3(16_{6,10}-15_{5,11})$A & 104 & 2.4 \\
1205: 347.599 & HCOOCH$_3(16_{6,10}-15_{5,11})$E & 105 & 1.7 \\
1206: 347.617 & HCOOCH$_3(28_{10,19}-27_{10,18})$A & 307 & 2.6 \\
1207: 347.628 & HCOOCH$_3(28_{10,19}-27_{10,18})$E & 307 & 3.6 \\
1208: 347.667 & UL & & 4.5 \\
1209: 347.759 & CH$_2$CHCN$(36_{2,34}-35_{2,32})$ & 317 & 8.3 & 0.7 & & \\
1210: 347.792 & UL & & 5.4 & 0.7 & & \\
1211: 347.842 & UL, $^{13}$CH$_3$OH & & 3.1 & 0.5 & & \\
1212: 347.916 & C$_2$H$_5$OH$(20_{4,17}-19_{4,16})$ & 251 & 3.3 & 0.7 & & \\
1213: 347.983 & UL & & & 0.6 & & \\
1214: 348.050 & HCOOCH$_3(28_{4,24}-27_{4,23})$E & 266 & 2.8 \\
1215: 348.066 & HCOOCH$_3(28_{6,23}-27_{6,22})$A & 266 & 3.0 \\
1216: 348.118 & $^{34}$SO$_2(19_{4,16}-19_{3,17})$ & 213 & 5.8 \\
1217: 348.261 & CH$_3$CH$_2$CN$(39_{2,37}-38_{2,36})$ & 344 & 11.2 & 1.2 & & \\
1218: 348.340 & HN$^{13}$C(4--3) & 42 & 16.1$^b$ & 2.0$^b$ & &\\
1219: 348.345 & CH$_3$CH$_2$CN$(40_{2,39}-39_{2,38})$ & 351 & 16.1$^b$ & 2.0$^b$ & & \\
1220: 348.388 & SO$_2(24_{2,22}-23_{3,21})$ & 293 & 9.3 & 0.5 & 1.0 & \\
1221: 348.518 & UL, HNOS$(1_{1,1}-2_{0,2})$ & & 10.6 & 0.7 \\
1222: 348.532 & H$_2$CS$(10_{1,9}-9_{1,8})$ & 105 & 7.4 & 1.9 & & \\
1223: 348.553 & CH$_3$CH$_2$CN$(40_{1,39}-39_{1,38})$ & 351 & 20.1 \\
1224: 348.910 & HCOOCH$_3(28_{9,20}-27_{9,19})$E & 295 & 11.0$^b$ & 1.6$^b$ & & \\
1225: 348.911 & CH$_3$CN$(19_{9}-18_{9})$ & 745 & 11.0$^b$ & 1.6$^b$ & & \\
1226: 348.991 & CH$_2$CHCN$(37_{1,36}-36_{1,35})$ & 325 & 5.9 \\
1227: 349.025 & CH$_3$CN$(19_{8}-18_{8})$ & 624 & 9.5 & 1.1 \\
1228: 349.107 & CH$_3$OH$(14_{1,13}-14_{0,14})$ & 43 & 12.2 & 3.1 & 1.3 & 1.1\\
1229: \hline \hline
1230: \multicolumn{7}{l}{\footnotesize $^a$ Doubtful detection since other close $v_t=2$ lines with similar upper }\\
1231: \multicolumn{7}{l}{\footnotesize energy levels were not detected.}\\
1232: \multicolumn{7}{l}{\footnotesize $^b$ Line blend.}\\
1233: \multicolumn{7}{l}{\footnotesize $^c$ No flux measurement possible because averaged over the given 5700\,AU negative }\\
1234: \multicolumn{7}{l}{\footnotesize features due to missing short spacings overwhelm the positive features (see Figs.~\ref{ch3oh_sample} \& \ref{so2_sample}).}\\
1235: \multicolumn{7}{l}{\footnotesize $^d$ Peak flux corrupted by neighboring SO$_2$ line.}\\
1236: \multicolumn{7}{l}{\footnotesize $^e$ Only detectable with higher
1237: spatial resolution \citep{beuther2005a}.}\label{linelistall}
1238: \end{longtable}
1239:
1240:
1241: \begin{table}[htb]
1242: \caption{Molecular column densities.}
1243: \begin{tabular}{lrrrr}
1244: \hline
1245: \hline
1246: & Orion-KL$^a$ & G29.96 & 23151 & 05358 \\
1247: \hline
1248: CH$_3$OH & $2{\sf x}10^{16}$ & $4{\sf x}10^{17}\,^b$ & $3{\sf x}10^{16}-1{\sf x}10^{17}\,^c$ & $4{\sf x}10^{15}-4{\sf x}10^{18}\,^d$ \\
1249: CH$_3$CH$_2$CN & $5{\sf x}10^{15}$ & $1{\sf x}10^{16}$ & -- & -- \\
1250: CH$_2$CHCN & $5{\sf x}10^{15}$ & $7{\sf x}10^{15}$ & -- & -- \\
1251: C$^{34}$S & $2{\sf x}10^{14}$ & $2{\sf x}10^{15}$ & $2{\sf x}10^{14}$ & $6{\sf x}10^{13}$ \\
1252: CH$_3$OCH$_3$ & $1{\sf x}10^{16}$ & $2{\sf x}10^{17}\,^e$ & -- & -- \\
1253: HCOOCH$_3$ & $6{\sf x}10^{15}$ & $8{\sf x}10^{16}$ & -- & -- \\
1254: $^{34}$SO & $6{\sf x}10^{14}$ & $1{\sf x}10^{16}$ & $4{\sf x}10^{15}$ & -- \\
1255: SO$_2$ & $2{\sf x}10^{15}$ & $3{\sf x}10^{16}$ & $3{\sf x}10^{16}$ & $1{\sf x}10^{16}\,^d$ \\
1256: HC$_3$N & $6{\sf x}10^{14}$ & $2{\sf x}10^{15}$ & -- & -- \\
1257: HN$^{13}$C & blend & blend & -- & -- \\
1258: H$_2$CS & $4{\sf x}10^{14}$ & $2{\sf x}10^{16}$ & -- & $4{\sf x}10^{14}$ \\
1259: C$_2$H$_5$OH & $2{\sf x}10^{15}$ & $6{\sf x}10^{16}$ & -- & -- \\
1260: SiO & $2{\sf x}10^{14}$ & $4{\sf x}10^{14}$ & $2{\sf x}10^{14}$ & -- \\
1261: CH$_3$CN & $2{\sf x}10^{15}$ & $1{\sf x}10^{16}$ & -- & $8{\sf x}10^{16}\,^d$ \\
1262: \hline
1263: \hline
1264: \end{tabular}
1265: \footnotesize{~\\
1266: $^a$ Calculated for lower average $T$ of 200\,K because of smoothing to 5700\,AU resolution (Figs.~\ref{sample_spectra} \& \ref{ch3oh_sample}). The source size was approximated by half the spatial resolution. The on average lower Orion-KL column densities are likely due to the largest amount of missing flux for the closest source of the sample.\\
1267: $^b$ From \citet{beuther2007d}. \\
1268: $^c$ From \citet{beuther2007f} for different sub-sources.\\
1269: $^d$ From \citet{leurini2007} for different sub-sources.\\
1270: $^e$ At lower temperature of 100\,K, because otherwise different lines would get excited.
1271: }
1272: \label{column}
1273: \end{table}
1274:
1275: \end{document}
1276:
1277: %%
1278: %% End of file `sample.tex'.
1279: