1: %% manuscript produces a one-column, double-spaced document:
2: %% preprint2 produces a double-column, single-spaced document:
3:
4: \documentclass[12pt,preprint]{aastex}
5: %\documentclass[manuscript]{aastex}
6: %\documentclass[preprint2]{aastex}
7:
8: \newcommand{\vdag}{(v)^\dagger}
9: \newcommand{\myemail}{skywalker@galaxy.far.far.away}
10:
11: \newcommand{\OIII}{O\,{\sc iii}}
12: \newcommand{\CII}{C\,{\sc ii}}
13: \newcommand{\Viv}{V\,{\sc iv}}
14: \newcommand{\Viii}{V\,{\sc iii}}
15: \newcommand{\Tiiii}{Ti\,{\sc iii}}
16: \newcommand{\Hi}{H\,{\sc i }}
17: \newcommand{\Hii}{H\,{\sc ii }}
18: \newcommand{\Tiii}{Ti\,{\sc ii}}
19: \newcommand{\Scii}{Sc\,{\sc ii}}
20: \newcommand{\Vii}{V\,{\sc ii}}
21: \def\Msun{{\hbox {M$_\odot$}}}
22:
23:
24: %% You can insert a short comment on the title page using the command below.
25:
26: \slugcomment{}
27:
28:
29: \shorttitle{New UfIB in NGC~7027}
30: \shortauthors{Goicoechea et al.}
31:
32:
33: \begin{document}
34:
35:
36: \title{A new Unidentified Far Infrared Band in NGC~7027}
37:
38:
39: \author{Javier R. Goicoechea, Jos\'e Cernicharo, Helena
40: Masso and Mar\'{\i}a Luisa Senent}
41: \affil{Departamento de Astrof\'{\i}sica Molecular e Infrarroja,
42: Instituto de Estructura de la Materia, CSIC,
43: Serrano 121, E--28006 Madrid, Spain.}
44: \email{javier@damir.iem.csic.es, cerni@damir.iem.csic.es,
45: masso@damir.iem.csic.es,
46: imts420@iem.cfmac.csic.es}
47:
48: \begin{abstract}
49:
50: We report on the detection of a molecular band centered at $\sim$98~$\mu$m
51: ($\sim$102~cm$^{-1}$), observed with the $\textit
52: {Infrared Space Observatory}$$^1$
53: in the young Planetary Nebula NGC~7027. The band structure and intensity
54: can not be reproduced by atomic fine structure lines, recombination
55: lines or by
56: the rotational emission of abundant molecules.
57: We discuss the possible contribution of the
58: low--energy bending modes of pure carbon chains to the
59: unidentified far--IR bands observed in $C$--rich evolved objects.
60: In particular, we speculate that the band emission
61: could arise from the
62: $\nu_9$ and $\nu_7$ bending modes of C$_6$ and C$_5$, for
63: which wavenumbers of 90$\pm$50 and 107$\pm$5~cm$^{-1}$ have been
64: estimated from photoelectron spectroscopy.
65:
66:
67: \end{abstract}
68:
69:
70:
71:
72: \keywords{infrared: stars --- circumstellar matter --- stars: individual
73: (NGC 7027) --- ISM: molecules --- line: identification}
74:
75:
76:
77: \footnotetext[1]
78: {Based on observations with ISO,
79: an ESA project with instruments funded by ESA Member States
80: (especially the PI countries: France, Germany, the Netherlands
81: and the United Kingdom) and with participation of ISAS and NASA.}
82:
83:
84: \section{Introduction}
85:
86: Carbon chains molecules are of special interest for Astrophysics
87: since Douglas (1977) proposed them as possible carriers
88: of the diffuse interstellar bands (DIBs).
89: In the past years, many polar carbon chains have been detected in
90: the interstellar and circumstellar media (ISM and CSM)
91: through their pure rotational spectrum at radio wavelengths.
92: Major examples are the cyanopolyynes, HC$_{2n+1}$N with $n$=1--5
93: (Turner 1971, Avery et al. 1976, Kroto et al. 1978, Broten et al. 1978,
94: Bell et al. 1997), and the hydrogenated carbon chain radicals such
95: as C$_5$H, C$_6$H, C$_7$H and C$_8$H (Cernicharo et al. 1986,
96: Gu\'elin et al. 1987, 1997, Cernicharo \& Gu\'elin 1996).
97:
98:
99: On the other hand, larger carbon molecular complexes such as the
100: polyatomic aromatic hydrocarbons (PAHs) are believed to
101: dominate the ubiquitous mid--IR emission seen in the unidentified
102: infrared bands (UIBs). However, only the single aromatic species,
103: benzene, has been so far identified through its IR active modes
104: (Cernicharo et al. 2001).
105:
106: All this chemistry richness suggests that the growth mechanisms
107: producing complex carbon molecules such as the PAHs or the fullerenes
108: are highly efficient. Nevertheless, the set of `building blocks''
109: and the possible chemical reactions still have to be identified and
110: clarified.
111: Among the possible species, the pure carbon chains,
112: (C$_n$; Van Orden \& Saykally 1998), could be
113: the `skeletons'' from which larger organic molecules can be formed
114: (see Cernicharo, Goicoechea \& Caux 2000, Cernicharo 2004).
115: Due to the lack of permanent electric dipole, these species
116: do not have rotational spectrum to be observed from radio telescopes.
117: The only way to detect them
118: in the dense ISM and CSM is through their asymmetrical stretching
119: modes around $\sim$5~$\mu$m (2000~cm$^{-1}$) and/or through their
120: low--energy bending modes around
121: $\sim$100~$\mu$m ($\sim$100~cm$^{-1}$).
122: The first technique allowed the detection of C$_3$ (Hinkle et al.
123: 1988) and C$_5$ (Bernath et al. 1989) in the circumstellar
124: envelope of the bright IR evolved star IRC+10216.
125: However, few sources (specially in the ISM) have
126: enough flux at $\sim$5~$\mu$m to allow systematic studies of
127: the C$_n$ stretching modes, and the less known far--IR bending modes are
128: the only way to detect these species.
129:
130: Before the launch of ISO (Kessler et al. 1996), we proposed to observe
131: the bending modes of several polyatomic molecules in the far--IR.
132: As a result, the $\nu_2$ bending mode of C$_3$ has been observed
133: in Sgr~B2 and IRC+10216 (Cernicharo et al. 2000).
134: In addition, we have detected several Unidentified far-IR bands (UfIBs) in
135: the ISO spectrum of many C--rich sources that may also be related
136: with the bending modes of carbon chains species. In particular
137: we have tentatively assigned a UfIB at $\sim$57.5~$\mu$m
138: ($\sim$174~cm$^{-1}$), observed in Sgr~B2 and
139: in $C$--rich evolved stars including the planetary nebula (PNe) NGC~7027,
140: to the $\nu_5$ bending mode of C$_4$ (Cernicharo, Goicoechea \& Benilan 2002).
141:
142: NGC~7027 is one of the most studied evolved objects in the Galaxy.
143: It is a young PNe, $\sim$1000~years since it left the AGB stage,
144: characterized by a relatively compact ionized region (Volk \& Kwok 1997)
145: driven by the radiation that arises from a central source at
146: $\sim$200,000~K (Latter et al. 2000).
147: In addition, a rich $C$--chemistry is taking place in the larger molecular
148: envelope ($\sim$40$''$ in CO; Masson et al. 1985) that surrounds
149: the inner photon dominated regions. In fact, NGC~7027 was the first object
150: to show UIB emission latter attributed to the PAH emission
151: (Gillett et al. 1973) and the only source where pure rotational lines
152: of CH$^+$ have been detected (Cernicharo et al. 1997).
153:
154: Therefore, the detection of unidentified features in the
155: IR spectrum of NGC~7027 has historically contributed to a better
156: understanding of the interstellar carbon complexity.
157: In this work, we report the detection of a new UfIB at
158: $\sim$98~$\mu$m ($\sim$102~cm$^{-1}$) in the far--IR spectrum
159: of NGC~7027 and speculate about it possible origin.
160:
161:
162: \section{Observations}
163:
164: The possible emission/absorption produced by the low--energy bending modes of
165: C$_n$ species has been searched using the \textit{Long--Wavelength
166: Spectrometer} (LWS; Clegg et al. Swinyard et al. 1996).
167: In the case of the NGC~7027 PNe, we have used all the
168: LWS AOT L01 data taken by ISO (see Herpin et al. 2002).
169: The LWS grating spectra of NGC~7027 taken during orbits
170: 21, 342, 349, 356, 363, 377, 537, 552, 559, 566, 579, 587, 594,
171: 601, 706, 713, 720, 727, 734, 741, 755, 762, 769, 776, and 783 have
172: been averaged.
173: The total on-source time was 53409~s and the signal--to--noise (S/N) ratio
174: is very high. Here we present part of the spectrum taken with the LWS
175: between $\sim$86 and 110~$\mu$m ($\sim$116--91~cm$^{-1}$) at
176: a resolution of $\lambda/\Delta\lambda\simeq200$.
177: The resulting spectrum is shown in Fig.~1, \textit{middle panel}.
178: We also present the same wavelength range in the LWS spectrum
179: [TDT19800158] of IRC+10216 (see Fig.~1, \textit{lower panel}). The full spectrum
180: was shown and modeled by Cernicharo et al. 1996.
181:
182: All data were processed following
183: pipeline number 9 and analyzed using the ISO Spectral Analysis
184: Package$^2$ (ISAP). Typical routines include: deglitching spikes due
185: to cosmic rays, oversampling and averaging individual scans and
186: removing baselines polynomials.
187:
188: \footnotetext[2]
189: {The ISO Spectral Analysis Package (ISAP) is a joint development by the LWS
190: and SWS Instruments Teams and Data Centers. Contributing institutes are CESR, IAS,
191: IPAC,MPE, RAL, and SRON.}
192:
193: In addition, we present part of the \textit{Short--Wavelength
194: Spectrometer} (SWS; de Graauw et al. 1996) spectrum of NGC~7027 between
195: $\sim$4--7~$\mu$m ($\sim$2500-1430~cm$^{-1}$).
196: The resolution of the SWS01 observation template in this interval is
197: $\lambda/\Delta\lambda\simeq1500$.
198: Many carbon chains (polar and non-polar) possess prominent IR active
199: stretching modes in this wavelength range.
200: However, Fig.~2 shows that the bulk of the emission
201: in NGC~7027 arises from narrow ionic and recombination lines from the
202: inner \Hii region. Only the well--known UIB broad emission at
203: $\sim$6.2~$\mu$m
204: is detected.
205: This band corresponds to the relaxation of excited vibrational states
206: of aromatic species pumped by absorption of visible--UV
207: photons from the hot central star.
208: The full SWS spectrum has been presented by Bernard Salas et al. (2001).
209:
210:
211:
212: \section{Results}
213:
214: After identified and modeled the molecular rotational
215: lines of CO $J$=30--29 to 25--24, CH$^+$ $J$=4--3 and OH ${^2}\Pi_{1/2}$
216: $J$=5/2--3/2 (see Herpin et al. 2002 for the details), a
217: strong band--like feature constituted by several lines
218: was identified between 91 and 102~$\mu$m.
219: Only the high--$J$ CO rotational emission produces modest
220: contribution in the band wavelengths. Fig.~1 also shows the
221: maximum contribution expected for the rotational $^{2}\Pi_{1/2}$
222: 5/2--3/2 line of OH at $\sim$98~$\mu$m estimated from the clearer
223: detections at $\sim$79 and $\sim$84~$\mu$m. Note that no water
224: lines have been detected in the far--IR spectrum of NGC~7027
225: (Cernicharo et al. 1997), and thus,
226: no more $O$-bearing species are contaminating the $\sim$86--110~$\mu$m
227: window.
228:
229: Due to the presence of an \Hii region in the inner envelope, we have
230: also searched for all the possible atomic
231: transitions that could arise from the nebular gas.
232: The most likely atomic lines in a PNe such as NGC~7027
233: (fine structure and recombination lines), taken from the
234: \textit{Atomic Line List} by Peter van Hoof have been analyzed.
235: We note that lines from [\Viv], [\Viii] and [\Tiiii],
236: with transitions in the considered wavelength range, have been previously
237: detected in the optical spectrum of NGC~7027 (Baluteau et al. 1995),
238: while several \Hi recombination lines are observed in the mid--IR
239: (e.g. see Fig.~2).
240:
241: To estimate the \Hi emission in the far--IR we have
242: considered a \Hii region ($2^{''}-4^{''}$) of
243: $\sim$0.022~\Msun (see the model of Volk \& Kwok 1997)
244: with $n_H$=$n_e$ and assumed that the \Hi population is given
245: by the Saha--Boltzmann equation.
246: With these parameters, we can predict the \Hi recombination
247: lines observed in the mid--IR by Bernard Salas et al. (2001),
248: and estimate the opacity of the far--IR lines. We found that
249: only the H$n$$\alpha$ series with $n$ from 10 to 15 may produce
250: modest emission in the LWS range. However, none of these lines
251: appear between 90 and 110~$\mu$m.
252: In addition, we have analyzed the possible overlapping with
253: fine structure lines. In particular, the
254: [\Tiii]96.68, 106.27 and [\Vii]97.79~$\mu$m lines
255: could produce some emission in the considered range.
256: Even assuming a vanadium abundance 10 times larger than the
257: solar abundance, the [\Vii]97.79~$\mu$m ($^{5}$D$_3$-$^{5}$D$_2$) line would be
258: extremely weak.
259: Note that the [\Vii]141.68~$\mu$m ($^{5}$D$_2$-$^{5}$D$_1$) line is
260: neither detected in NGC~7027.
261: Finally, the solar abundance of Ti is $\sim$10 times that
262: of vanadium. We also computed that is too low to produce significant
263: emission in the [\Tiii]96.68, 106.27 lines.
264:
265:
266: Hence, the integrated band intensity and the different lines
267: or sub-bands could not
268: be fully assigned to any of these atomic lines nor to
269: the pure rotational lines with significant line
270: strength arising from the light species
271: in the circumstellar envelope.
272:
273: We also note that after subtracting the crowed emission of CO, HCN and HCN$_{vib}$
274: (pure rotational lines in vibrational excited states) from the spectrum of
275: IRC+10216, some of the remaining peaks
276: also agreed with those unidentified peaks observed in NGC~7027
277: (see Fig. 1). However, the spectral resolution is too poor
278: to distinguish unidentified features from the forest of
279: HCN$_{vib}$ rotational lines that contaminate the far--IR spectrum
280: of IRC+10216 but are missing in NGC~7027.
281: The HCN abundance in IRC+10216 is very high, $\sim$3$\times$10$^5$,
282: with HCN/CO=1/10. In fact, HCN is the main coolant of this $C$--rich
283: AGB star (Cernicharo et al. 1999). However, the molecule is
284: photodissociated during the post--AGB evolution as the UV radiation field from
285: the evolving star increases. From far--IR observations of HCN pure rotational
286: lines, Herpin et al. (2002) found minimum abundances of HCN/CO=1/100 and 1/1000
287: for the proto--PNe CRL~2688 and CRL~618 respectively.
288: The extreme case is NGC~7027, were a great part of
289: HCN molecules must have been photodissociatd in CN and H.
290: In fact, Bachiller et al. (1997) found CN/HCN$\sim$10 for this object. Thus,
291: the low abundance of HCN compared to IRC+10216 is consistent
292: with the non detection of far--IR HCN lines (Herpin et al. 2002).
293: Pure rotational lines of other cyanopolyynes species such as HC$_3$N
294: can not produce emission in this range because of the high energy
295: levels associated with its far--IR transitions.
296:
297:
298: Taking into account the high S/N
299: ratio of the NGC~7027 spectrum, and the absence of the band
300: in sources such as Orion, Sgr~A or $O$--rich evolved stars, we believe
301: that the lines are real.
302: Due to the chemistry of NGC~7027,
303: the most probable carrier should be the low--energy bending mode of a
304: polyatomic molecule containing carbon.
305:
306: \section{Discussion}
307:
308: Both PAHs and pure carbon chains possesses low--energy bending modes in the
309: far--IR. Contrary to the UIB emission in the mid--IR, the far--IR skeletal modes
310: of PAHs (vibrations associated with the bending of the skeletal structure)
311: depends of the exact nature of the species.
312: Hence, the rovibrational structure of the PAH skeletal mode has to be calculated
313: for each specific PAH. As a example, the lowest energy active
314: mode of coronene (C$_{24}$H$_{12}$) lies at $\sim$127~cm$^{-1}$
315: (Joblin et al. 2002), while for ovalene
316: (C$_{32}$H$_{14}$), two far--IR skeletal modes at $\sim$120~cm$^{-1}$
317: and $\sim$66~cm$^{-1}$ have been theoretically investigated
318: (Mulas et al. 2003). The far-IR emission of a few PAHs has also
319: been observed in gas phase experiments (Zhang et al. 1996).
320: However, if one of the observed UfIBs belongs to a specific PAHs, other
321: bands (generally stronger) should be observed in the IR spectrum.
322: This is not the case of the new
323: UfIB presented in this work and we have considered the low--energy modes
324: of pure carbon chains as possible carriers of the observed UfIB.
325:
326: The detection of several rovibrational lines of the $\nu_2$
327: bending mode of C$_3$ ($\sim$63~cm$^{-1}$) in Sgr~B2 and
328: IRC+10216 evidenced high abundances for this species
329: and opened the possibility to detect more carbon chains
330: in the ISM and CSM (Cernicharo et al. 2000).
331: At the low resolution of the LWS/grating, the more intense
332: C$_3$ lines [Q(2,4,6)] are blended with the strong
333: [\CII]158~$\mu$m line and no assignation could be made
334: in NGC~7027.
335: However, the tentative detection of the $\nu_5$ bending
336: mode of C$_4$ in many objects including NGC~7027
337: ($\sim$174~cm$^{-1}$; Cernicharo et al. 2002),
338: if confirmed, will imply a C$_3$/C$_4$$<$10
339: abundance ratio, which will suggest that the abundance of C$_n$ does not
340: decrease drastically when $n$ increases.
341:
342: In spite of their importance, the C$_n$ bending modes are difficult
343: to characterize spectroscopically in the laboratory or
344: by \textit{ab initio} computations.
345: As expected, the main uncertainties difficulting their astronomical
346: detection are the band origin and the IR intensity.
347: Moreover, the induced dipole moments
348: and other spectroscopic constants
349: are often known for the fundamental transition but less or nothing
350: is known about overtone transitions within the bending mode or transitions
351: to other excited states including the stretching modes.
352:
353: The ground electronic state is different
354: for the odd-- and even-- numbered cumulenic chains. This property determines
355: the observed band shape of the bending mode.
356: Odd--numbered linear C$_n$'s have a singlet
357: $^{1}\Sigma^{+}_{g}$ ground electronic state which results in
358: a perpendicular $^{1}\Pi_{vib}-{^1}\Sigma_{vib}$ vibronic spectrum with a
359: strong --$Q$ branch and weak --$P$ and --$R$ branches.
360: On the other hand, even--numbered linear C$_n$'s have a triplet
361: $^{3}\Sigma^{-}_{g}$ ground electronic state. Hence, each rotational
362: line is split in three components
363: due to spin couplings. This splitting
364: increase with the number of atoms but is particularly
365: small for C$_4$ and C$_6$ (Giesen et al. 2001).
366: The vibronic transition is now $^{3}\Pi_{vib}-{^3}\Sigma_{vib}$, and
367: despite the null component of the electronic orbital
368: momentum in the ground state, the spin--orbit constant (A$_{SO}$)
369: could be large in even-- C$_n$'s.
370: In such a case, the resulting band--shape will
371: be notoriously affected by the value of A$_{SO}$.
372:
373: \subsection{Tentative detection of C$_6$ and C$_5$}
374:
375: With the exception of the $\nu_2$ mode of C$_3$, the band origins of
376: other C$_n$ bending modes are not accurately constrained.
377: Hence, their assignation as carriers of the UfIBs
378: is not obvious. Although the low--lying bending modes of C$_5$ and C$_6$
379: have not been directly observed in gas phase, different studies suggest wavenumbers
380: around $\sim$100~cm$^{-1}$ (see below).
381: The observed new UfIB around $\sim$98~$\mu$m
382: is composed by several peaks at
383: $\sim$91.8, 92.8, 93.9, 95.2, 97.6, 98.9, 100.5 and 101.7~$\mu$m.
384: In the following, we will consider the possibility that
385: \textit{any of these peaks is related with the
386: $\nu_9$ and $\nu_7$ bending modes of C$_6$ and C$_5$ respectively}.
387:
388:
389: C$_5$ was first detected in the gas phase by Heath et al. (1989)
390: who observed the $\nu_3$ stretching mode at $\sim$2169~cm$^{-1}$
391: ($\sim$4.6~$\mu$m). The same mode was observed in
392: IRC+10216 by Bernath et al. (1989). Actually, C$_5$ is the largest
393: C$_n$ detected in the CSM.
394: On the other hand, C$_6$ was first identified in the laboratory by its
395: electronic spin resonance spectrum in an Ar matrix (Van Zee et al. 1987)
396: and later observed in gas phase through its $\nu_4$ stretching mode at
397: $\sim$1960~cm$^{-1}$ ($\sim$5.1~$\mu$m) by Hwang et al. (1993).
398: Moazzen--Ahmadi et al.
399: (1989) observed the ($\nu_3$+$\nu_7$)--$\nu_7$ and ($\nu_3$+2$\nu_7$)--2$\nu_7$
400: hot bands arising from the $\nu_7$ bending mode of C$_5$.
401: From the $l$--doubling constant $q_7$, they estimated a frequency of
402: $\nu_7$=118$\pm$3~cm$^{-1}$.
403: More recently, C$_5$ has been observed in photoelectron spectra.
404: In particular, Arnold et al. (1991) measured the 2$\nu_7$ transition
405: and estimated $\nu_7$=101$\pm$45~cm$^{-1}$, while Kitsopoulos et al.
406: (1991) obtained $\nu_7$=107$\pm$5~cm$^{-1}$. \textit{Ab initio}
407: calculations predict an infrared intensity around 36~km~mol$^{-1}$
408: [$A$($\nu_7$=1-0)$\simeq$0.06~s$^{-1}$; Hutter \& L\"uthi 1994; Martin et al.
409: 1995]. Fig.~1 shows the expected band--shape for a
410: $^{1}\Pi_{vib}-{^1}\Sigma_{vib}$ transition with this intensity and the molecular
411: constants for the $\nu_7$ mode (Moazzen--Ahmadi et al.)
412: centered at 104.8~cm$^{-1}$ (95.4~$\mu$m). As said before, any
413: of the band peaks observed in NGC~7027 around
414: $\sim$98~$\mu$m ($\sim$102~cm$^{-1}$) could well be responsible
415: of the $\nu_7$ mode. However, only the best fit to the
416: band including the C$_6$ model
417: is shown in Fig.~1.
418: If the carrier of the 95.5~$\mu$m feature is finally C$_5$
419: at $\nu_7$=104.8~cm$^{-1}$,
420: the estimated column density
421: assuming an excitation temperature
422: of 100~K (typical of a PNe envelope) is
423: $N$(C$_5$)=1.8$\times$10$^{14}$~cm$^{-2}$.
424: Assuming 1.5--4.0 magnitudes of visual extinction in the neutral envelope
425: (Hasegawa et al. 2000), the typical abundance of C$_5$ in NGC~7027
426: would be (0.5--1.0)$\times$10$^{-7}$, similar to that derived by Bernath
427: et al. (1989) in IRC+10216.
428:
429: Fig.~2 shows that, for the same column densities required to
430: reproduce the bending modes, the $\nu_4$ and $\nu_3$ stretching
431: modes of C$_6$ and C$_5$ are not detected in NGC~7027.
432: Different excitation conditions of the stretching and bending
433: modes and geometrical effects may explain this conjecture.
434: Although carbon clusters have a moderate
435: size, we could also expect an efficient pumping of the stretching and
436: bending modes through UV photons. Mid--IR photons could also contribute
437: to the excitation of the high--$\nu_{bending}$
438: excited states but also of the stretching modes. Finally, the low--energy bending
439: modes could be pumped alone by the absorption of far--IR dust photons
440: (see the case of the C$_3$ excitation in Cernicharo et al. 2000).
441: However, the smaller volume of the photon--dominated regions relative to
442: the far--IR dusty envelope in NGC~7027, and the larger mid--IR flux of
443: IRC+10216 (where the C$_3$ and C$_5$ stretching modes were detected)
444: compared to NGC~7027 (a factor $\sim$10$^3$ at 5~$\mu$m, but only a factor
445: $\sim$4 at 100~$\mu$m), can favor the observation of the C$_n$ bending
446: modes within the large ISO beam.
447: Under these conditions, the detection of the bending modes and the
448: non--detection of the stretching modes in NGC~7027 can be plausible.\\
449:
450:
451:
452:
453:
454: The $\nu_9$ low--energy bending mode of C$_6$ is even less known.
455: \textit{Ab inito} calculations predict a $\nu_9$$\sim$108~cm$^{-1}$
456: frequency and an infrared intensity of $\sim$25~km~mol$^{-1}$
457: [$A$($\nu_9=1-0$)$\simeq$0.035~s$^{-1}$; Martin et al. 1990; 1995), while
458: photoelectron spectroscopy experiments estimate
459: $\nu_9$=97$\pm$45~cm$^{-1}$ (Arnold et al. 1991) and
460: $\nu_9$=90$\pm$50~cm$^{-1}$ (Xu et al. 1997).
461: Theoretical predictions have to be taken into account with caution
462: since they rely in the harmonic approximation which can be in error
463: for large amplitude benders such as radicals with low-lying
464: electronic states. In particular, the infrared intensity has to be
465: considered as a lower limit (and column densities as upper limits).
466: \textit{Ab initio} calculations do predict a low-lying electronic
467: state for open--shell species such as C$_4$ or C$_6$. The term value for the
468: $^{1}\Delta_g$ lowest energy excited state of C$_6$ has been recently
469: determined by Xu et al. only at $\sim$1400~cm$^{-1}$ from the ground.
470: Hence, A$_{SO}$ can effectively be large even in the ground
471: state ($\Lambda$=0).
472: As an example,
473: the A$_{SO}$ constant of C$_4$H ($^{2}\Sigma$ ground state) is
474: very large, $\simeq$3~cm$^{-1}$ (Yamamoto et al. 1987), because its
475: $^{2}\Pi$ electronic excited state is only $\simeq$468~cm$^{-1}$
476: above the ground. In the case of even--numbered C$_n$ chains,
477: A$_{SO}$ can easily be larger because of the higher spin multiplicity
478: of the ground state and the larger $\Lambda$ value
479: of the lowest--lying electronic excited state.
480: Fig.~1 also shows the expected band--shape for a
481: $^{3}\Pi_{vib}-{^3}\Sigma_{vib}$ transition with A$_{SO}$=3.5~cm$^{-1}$
482: (we estimated A$_{SO}$$\sim$4~cm$^{-1}$ for C$_4$),
483: C$_6$ molecular constants from Hwang et al. (1993) and van
484: Zee et al. (1987;1988), and the $\nu_9$ band origin at
485: 98.3~cm$^{-1}$ (101.7~$\mu$m). Assuming the same excitation temperature
486: than for C$_5$ we derive $N$(C$_6$)=0.8$\times$10$^{14}$~cm$^{-2}$.
487: Hence, the C$_6$ abundance will be a factor 2 smaller
488: than that of C$_5$.
489:
490: Much more work has to be done to fully understand and characterize
491: the low--energy vibrations of C$_n$ chains in order to assign the UfIBs
492: observed by ISO. The cyclic C$_n$ isomers could also produce
493: spectral features in the far--IR, however, their active modes are even
494: less known.
495: Waiting for such progresses, space observations offer the unique
496: opportunity to obtain spectra in the far--IR domain where the bending
497: modes appear and to motivate further laboratory and theoretical studies.
498: The presence of small C$_n$ chains in the space
499: as well as their high reactivity, suggest that these species can be
500: involved in the formation of more complex organic molecules.
501: In fact, for cumulenic clusters with $n$=10 to 20, linear structures
502: are thought to close into rings, while for $n$$>$30,
503: these species are thought to be more stable in aromatic and fullerene--like
504: structures (O' Brien et al. 1987).
505:
506: Future space heterodyne telescopes, such as the \textit{Herschel Space
507: Observatory}, with much better sensitivity and
508: spectral resolution in the far--IR,
509: should allow the detection of longer C$_n$ chains through their
510: low--energy bending modes. This will be the fingerprint needed to
511: understand the formation and the nature of the UIBs and UfIBs carriers.
512:
513: \clearpage
514:
515: \acknowledgments
516:
517: We thank Spanish DGES and PNIE for funding
518: support under grants PANAYA2000-1784, ESP2001-4516, AYA2002-10113-E,
519: ESP2002-01627, AYA2002--02117 and AYA2003-02785-E.
520: We also thank Ana Heras for providing us the processed SWS
521: spectrum of NGC~7027 and F.~Najarro and C.~Joblin for useful comments about
522: atomic lines and PAHs respectively. We have used the Atomic Line List v2.04
523: of Peter van Hoof (http://www.pa.uky.edu/ $\sim$peter/atomic).
524:
525:
526:
527: \begin{thebibliography}{}
528:
529:
530: \bibitem{Ar91}
531: Arnold, D. W., Bradforth, S. E., Kitsopoulos, T. N. \& Neumark, D. M.
532: 1991, JChPh, 95, 8753
533:
534: \bibitem{Ave76}
535: Avery, L. W., Broten, N. W., MacLeod, J. M., Oka, T., \& Kroto, H. W.
536: 1976, ApJ, 205, L173
537:
538: \bibitem{Bac97}
539: Bachiller, R., Forveille, T., Huggins, P. J., \& Cox, P.
540: 1997, A\&A, 324, 1123
541:
542:
543: \bibitem{Bal95}
544: Baluteau, J.-P., Zavagno, A., Morisset, C., \& Pequignot, D.
545: 1995, A\&A, 303, 175
546:
547: \bibitem{Bell87}
548: Bell, M. B., Feldman, P. A., Travers, M.J., McCarthy, M.C., Gottieb, C.A.
549: \& Thaddeus, P. 1997, ApJ, 483, L61
550:
551:
552: \bibitem{Ber96}
553: Bernard Salas, J., Pottasch, S. R., Beintema, D. A., \& Wesselius, P. R.
554: 2001, A\&A, 367, 949
555:
556: \bibitem{Ber89}
557: Bernath, P.F., Hinkle, K.H., \& Keady, J.J. 1989, Science, 244, 562
558:
559: \bibitem{Brot78}
560: Broten, N. W., Oka, T., Avery, L. W., MacLeod, J. M., \& Kroto, H. W.
561: 1978, ApJ, 223, L105
562:
563: \bibitem{Cer86}
564: Cernicharo, J., Kahane, C., G\'omez-Gonz\'alez, J., \& Gu\'elin, M.
565: 1986, A\&A, 164, L1
566:
567: \bibitem{Cer96}
568: Cernicharo, J. \& Gu\'elin, M. 1996, A\&A, 309, L27
569:
570:
571: \bibitem{Cer96b}
572: Cernicharo, J., Barlow, M. J., Gonzalez-Alfonso, E. et al.
573: 1996, A\&A, 315, L201
574:
575: \bibitem{Cer97}
576: Cernicharo, J., Liu, X.W., Gonz\'alez-Alfonso, E., Cox, P.,
577: Barlow, M. J., Lim, T., Swinyard, B. M. 1997, ApJ, 483, L65
578:
579:
580: \bibitem{Cer00}
581: Cernicharo, J., Goicoechea, J.R., \& Caux, E. 2000,
582: ApJ, 534, L199
583:
584: \bibitem{Cer01}
585: Cernicharo, J., Heras, A.M., Tielens, A. G. G. M., Pardo, J.R.,
586: Herpin, F. Guelin, M., Waters, L. B. F. M. 2001,
587: ApJ, 546, L123
588:
589:
590: \bibitem{Cer02}
591: Cernicharo, J., Goicoechea, J.R., \& Benilan, Y. 2002, ApJ, 580, L157
592:
593: \bibitem{Cer04}
594: Cernicharo, J. 2004, ApJ, in press.
595:
596: \bibitem{Cleg96}
597: Clegg, P. E., et al. 1996, A\&A, 315 L38
598:
599:
600: \bibitem{deG96}
601: de Graauw, T., et al. 1996, A\&A, 315, L49
602:
603:
604: \bibitem{Doug77}
605: Douglas, A.E. 1977, Nature, 269, 130
606:
607:
608: \bibitem{Gie01}
609: Giesen, T. F., Berndt, U., Yamada, K.M.T. et al. 2001b,
610: ChemPhysChem. 4, 242
611:
612: \bibitem{Gil73}
613: Gillett, F. C., Forrest, W. J., \& Merrill, K. M.
614: 1973, ApJ, 183, 87
615:
616: \bibitem{Gue87}
617: Gu\'elin, M., Cernicharo, J., Kahane, C., G\'omez-Gonz\'alez, J., \& Walmsley, C. M. 1987, A\&A, 175, L5
618:
619:
620: \bibitem{Gue97}
621: Gu\'elin, M., et al. 1997, A\&A, 317, L1
622:
623:
624: \bibitem{Has00}
625: Hasegawa, T., Volk, K., \& Kwok, S.
626: 2000, ApJ, 532, 994
627:
628: \bibitem{Hea89}
629: Heath, J. R., Cooksy, A. L., Gruebele, M. H. W.,
630: Schmuttenmaer, C. A., \& Saykally, R. J. 1989, Science, 244, 564
631:
632: \bibitem{Her2002}
633: Herpin, F., Goicoechea, J. R., Pardo, J. R., \& Cernicharo, J.
634: 2002, ApJ, 577, 961
635:
636: \bibitem{Hin88}
637: Hinkle, K.W., Keady, J.J., \& Bernath, P.F. 1988, Science, 241, 1319
638:
639: \bibitem{Hut94}
640: Hutter, J., \& L\"uthi, H.P., 1994, JChPh, 101, 2213
641:
642:
643: \bibitem{Hwa93}
644: Hwang, H.J., van Orden, A., Tanaka, K., Kuo, E.W., Heath, J.R.,
645: \& Saykally, R. J. 1993, JMolPhys, 79, 769
646:
647: \bibitem{Kes96}
648: Kessler, M. F., Steinz, J. A., Anderegg, M. E. et al.
649: 1996, A\&A, 315, L27
650:
651: \bibitem{Kit91}
652: Kitsopoulos, T. N., Chick, C. J., Zhao, Y., \& Neumark, D. M.
653: 1991, JChPh, 95, 5479
654:
655: \bibitem{Krot78}
656: Kroto, H. W., Kirby, C., Walton, D. R. M., Avery, L. W.,
657: Broten, N. W., MacLeod, J. M., \& Oka, T. 1978, ApJ, 219, L133
658:
659: \bibitem{Lat00}
660: Latter, W.B., Dayal, A., Bieging, J.H., Meakin, C,, Hora, J.L.,
661: Kelly, D.M. \& Tielens, A. G. G. M.
662: 2000, ApJ, 539, L783
663:
664:
665: \bibitem{Mar90}
666: Martin, J.M.L, Francois, J.P., \& Gijbels, R.
667: 1990, JChPh, 93, 8850
668:
669: \bibitem{Mar95}
670: Martin, J.M.L, El--Yazal, J., \& Francois, J.P.
671: 1995, ChemPhysLett, 242, 570
672:
673: \bibitem{Mas85}
674: Masson, C.R. et al.
675: 1985, ApJ, 292, 464
676:
677:
678:
679: \bibitem{Moa89}
680: Moazzen-Ahmadi, N., McKellar, A. R. W., \& Amano, T. 1989,
681: JChPh, 91, 2140
682:
683:
684: \bibitem{Obr87}
685: O'Brien, S.C., Heath, J.R., Curl, R.F., \& Smalley, R.E. 1987
686: JChPh, 88, 220
687:
688:
689: \bibitem{Swi96}
690: Swinyard, B. M., et al. 1996, A\&A, 315, L43
691:
692: \bibitem{Tur71}
693: Turner, B. E. 1971, ApJ, 163, L35
694:
695: \bibitem{Van98}
696: van Orden, A., \& Saykally, R.J. 1998, ChemRev, 98, 2313--2357
697:
698: \bibitem{Van87}
699: van Zee, R. J., Ferrante, R. F., Zeringue, K. J., \& Weltner, W., Jr.
700: 1987, JChPh, 86, 5212
701:
702:
703: \bibitem{Van88}
704: van Zee, R. J., Ferrante, R. F., Zeringue, K. J., Weltner, W., Jr.,
705: \& Ewing, D. W. 1988, JChPh, 88, 3465
706:
707: \bibitem{Vol97}
708: Volk, K., \& Kwok, S. 1997, ApJ, 477, 722
709:
710:
711: \bibitem{Xu97}
712: Xu, C., Burton, G.R., Taylor, T.R., \& Neumark, D.M.
713: 1997, JChPh, 107, 3428
714:
715: \bibitem{Yam87}
716: Yamamoto, S., Saito, S., Guelin, M., Cernicharo, J., Suzuki, H., \&
717: Ohishi, M. 1987, ApJ, 323, L149
718:
719: \bibitem{Zhan96}
720: Zhang, K., Guo, B., Colarusso P. \& Bernath, P.F.
721: 1996, Sci, 274, 582
722:
723:
724:
725:
726: \end{thebibliography}
727:
728:
729:
730: \begin{figure}
731: \caption
732: {\textit{Top panel}:
733: Expected band shape for a $^1\Pi_{vib}-{^1}\Sigma_{vib}$ transition
734: (as the $\nu_7$ mode of C$_5$) and
735: for a $^3\Pi_{vib}-{^3}\Sigma_{vib}$ transition (as the $\nu_9$ mode of C$_6$)
736: with A$_{SO}$=3.5~cm$^{-1}$. The excitation temperature for both
737: transitions is 100~K. The intrinsic line width is 10~km~s$^{-1}$
738: and the spectral resolution $\lambda/\Delta\lambda$ is 200.
739: ---\textit{Middle panel}: Observed ISO/LWS spectrum of NGC~7027
740: between $\sim$86 and 110~$\mu$m. The ordinate corresponds to the
741: flux over the continuum flux and the abscissa to the wavelength in $\mu$m.
742: Large Velocity Gradient model for the pure
743: rotational emission of CO, CH$^+$ and OH (Herpin et al. 2002) and
744: the total model for the bending modes of
745: C$_5$ and C$_6$.
746: The thick horizontal arrow represents
747: the observational uncertainty for the $\nu_7$ C$_5$
748: band origin (from Kitsopoulos et al. 1991). The experimental
749: uncertainty for the $\nu_9$ C$_6$ mode is 90$\pm$50~cm$^{-1}$
750: (Xu et al. 1997). See text.
751: ---\textit{Bottom panel}: Observed ISO/LWS spectrum of IRC+10216
752: between $\sim$86 and 110~$\mu$m. The ordinate corresponds to the
753: continuum subtracted flux and the abscissa to the wavelength in $\mu$m.
754: Large Velocity Gradient model for the pure
755: rotational emission of CO $\nu$=0, 1, $^{13}$CO $\nu$=0, HCN and H$^{13}$CN
756: $\nu=$0, and HCN$_{vib}$ $\nu_2$=1, 2 and $\nu_{1,3}$=1 (Cernicharo et al. 1996).
757: The rotational transitions of CO, HCN in the ground state, and
758: HCN$_{vib}$ $\nu_2$=1, 2 and $\nu_{1,3}$=1 are indicated.
759: }
760: \end{figure}
761: \plotone{f1_color.eps}
762:
763: \clearpage
764:
765: \begin{figure}
766: \caption{Observed ISO/SWS spectrum of NGC~7027
767: between $\sim$4 and 7~$\mu$m. The ordinate scale is F/F$_c$ and the
768: wavelength is in $\mu$m. The fine structure and recombination
769: lines from the ionized region are labeled.
770: The expected $\Sigma_{vib}-\Sigma_{vib}$ parallel spectra from the
771: $\nu_3$ C$_3$, $\nu_3$ C$_4$, $\nu_3$ C$_5$, $\nu_4$ C$_6$
772: and $\nu_4$ C$_7$ stretching modes are
773: also shown. The excitation temperature for all bands is also
774: 100~K. The intrinsic line width is 10~km~s$^{-1}$
775: and the spectral resolution $\lambda/\Delta\lambda$ is 1500.
776: Column densities are $N$(C$_5$)=1.8$\times$10$^{14}$~cm$^{-2}$
777: (from the $\nu_7$ bending mode), $N$(C$_6$)=0.8$\times$10$^{14}$~cm$^{-2}$
778: (from the $\nu_9$ bending mode), $N$(C$_3$)=10$\times$$N$(C$_5$),
779: $N$(C$_7$)=$N$(C$_5$)$/$10 and $N$(C$_4$)=0.8$\times$10$^{15}$~cm$^{-2}$
780: (from Cernicharo et al. 2002).}
781: \end{figure}
782: \plotone{f2_color.eps}
783:
784:
785:
786: \end{document}
787:
788:
789:
790: