0808.1341/ms.tex
1: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
2: %                         ASTRO-PH
3: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
4: %  17-6 version sent to co-authors giving them to the 1july.
5: %  18-6 received comments from M. Honda.
6: %  07-08 Apj editted version of 1st submission
7: %  08-08 Apj editted version of 2st submission
8: 
9: %%\documentclass[12pt,preprint]{aastex}
10: %\documentclass[manuscript]{aastex}
11: %\documentclass[preprint2]{aastex}
12: \documentclass[apj]{emulateapj}
13: %\documentclass[manuscript]{aastex}
14: %\documentclass[apjl]{emulateapj}
15: %\usepackage{natbib}
16: 
17: %\usepackage{apjfonts}
18: %\usepackage{mathptmx}
19: \usepackage{txfonts}
20: 
21: %\input epsf.sty
22: 
23: %\usepackage{graphicx}
24: %\usepackage{txfonts}
25: %\usepackage{natbib}
26: %\usepackage{aalongtable}
27: \bibpunct{(}{)}{;}{a}{}{,}
28: \newcommand{\Msol}{\mbox{$M_{\odot}$}}
29: \newcommand{\Msunyr}{\mbox{$\rm M_{\odot}$\,yr$^{-1}$}}
30: \newcommand{\Lsun}{\mbox{$L_{\odot}$}}
31: \newcommand{\Rsun}{\mbox{$R_{\odot}$}}
32: \newcommand{\zsun}{\mbox{$Z_{\odot}$}}
33: \newcommand{\Teff}{\mbox{$T_{\rm eff}$}}
34: \newcommand{\IR}{IRC\,+10\,420~}
35: \newcommand{\Brg}{Br$\gamma$~}
36: \newcommand{\mdot}{\mbox{$\dot{M}$}}
37: \newcommand{\Mdot}{\mbox{$\dot{M}$}}
38: %\newcommand{\micron}{$\mu$m~}
39: 
40: \def\mathstacksym#1#2#3#4#5{\def#1{\mathrel{\hbox to 0pt{\lower#5\hbox{#3}\hss} \raise #4\hbox{#2}}}}
41: \mathstacksym\gta{$>$}{$\sim$}{1.5pt}{3.5pt} % greater than approximately
42: \mathstacksym\lta{$<$}{$\sim$}{1.5pt}{3.5pt} % less than approximatel
43:                                 %
44: \begin{document}
45: 
46: %%%%%%%%%%%%%%%%%%%%%%%% Title Page begins %%%%%%%%%%%%%%%%%%%%%%%%%%%%%
47: 
48: \title{A red supergiant nebula at 25 micron: arcsecond scale mass-loss asymmetries of $\mu$ Cep\altaffilmark{1}}
49: 
50: \altaffiltext{1}{Based on data collected at the Subaru telescope, which is operated by the National Astronomical Observatory of Japan.}
51:      
52:                            %
53: \shorttitle{Subaru imaging of $\mu$ Cep}
54: \shortauthors{de Wit et al.}
55:                                 %
56: \author{W.J. de Wit\altaffilmark{2}, R.D. Oudmaijer\altaffilmark{2}, T. Fujiyoshi\altaffilmark{3}, M.G. Hoare\altaffilmark{2}, M. Honda\altaffilmark{4},
57: H. Kataza\altaffilmark{5}, T. Miyata\altaffilmark{6}, Y. K. Okamoto\altaffilmark{7}, T. Onaka\altaffilmark{8}, S. Sako\altaffilmark{6}, T. Yamashita\altaffilmark{3}} 
58:                                 %
59: % hondamt@ir.isas.jaxa.jp
60: %
61: %
62: %
63: %\email{w.j.m.dewit@leeds.ac.uk}
64:                                 %
65: \altaffiltext{2}{School of Physics \& Astronomy, University of Leeds, Woodhouse Lane, Leeds LS2 9JT, UK; w.j.m.dewit@leeds.ac.uk}
66: \altaffiltext{3}{Subaru Telescope, National Astronomical Observatory of Japan, National Institutes of Natural Sciences, 650 North A'ohoku Place, Hilo, HI 96720, USA}
67: \altaffiltext{4}{Department of Information Science, Kanagawa University, 2946 Tsuchiya, Hiratsuka, Kanagawa, 259-1293, Japan}
68: \altaffiltext{5}{Department of Infrared Astrophysics, Institute of Space and Astronautical Science, Japan Aerospace Exploration Agency, Sagamihara, Kanagawa, 229-8510, Japan}
69: \altaffiltext{6}{Institute of Astronomy, University of Tokyo, Osawa 2-21-1, Mitaka, Tokyo 181-0015, Japan}
70: \altaffiltext{7}{Faculty of Science, Ibaraki University, 2-1-1 Bunkyo, Mito, Ibaraki, 310-8512, Japan}
71: \altaffiltext{8}{Department of Astronomy, Graduate School of Science, University of Tokyo, Bunkyo-ku, Tokyo 113-0022, Japan}
72: %
73: %\date{Received date; accepted date}
74:                                 %
75: \begin{abstract}
76: We present diffraction limited (0.6\arcsec) 24.5\,\micron~ Subaru/COMICS
77: images of the red supergiant $\mu$ Cep. We report the detection of a
78: circumstellar nebula, that was not detected at shorter wavelengths. It
79: extends to a radius of at least 6\arcsec~in the thermal infrared. On
80: these angular scales, the nebula is roughly spherical, in contrast, it
81: displays a pronounced asymmetric morphology closer in. We
82: simultaneously model the azimuthally averaged intensity profile of the
83: nebula and the observed spectral energy distribution (SED) using
84: spherical dust radiative transfer models. The models indicate a
85: constant mass-loss process over the past 1000 years, for mass-loss
86: rates a few times $10^{-7}$\,\Msunyr.  This work supports the idea that
87: at least part of the asymmetries in shells of evolved massive stars
88: and supernovae may be due to the mass-loss process in the red
89: supergiant phase.
90: \end{abstract}
91: %
92: \keywords{(stars:) supergiants - stars: evolution - stars : individual $\mu$ Cep - stars: mass-loss} 
93: 
94: %%%%%%%%%%%%%%%%%%%%%%%% Main Text begins %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
95: 
96: \section{Introduction}
97: \label{intro}
98: Although the final stages of the post-main sequence evolution of
99: massive stars do not last long, it is here where most of the mass is
100: lost and the shaping of the pre-supernova ejecta takes place.  Key in
101: this respect are the Red Supergiants (RSGs) which represent a phase in
102: the life of stars with initial masses 10--30\,M$_{\odot}$. During this
103: phase lasting 10$^4 -10^5$ years, the stars lose prodigious amounts of
104: mass at a rate of order 10$^{-6} - 10^{-4}$ M$_{\odot}$ yr$^{-1}$ (van Loon et
105: al. 2005\nocite{2005A&A...438..273V}; Massey et
106: al. 2008\nocite{2008arXiv0801.1806M}), and the final mass of the
107: objects is mostly set during this phase. RSGs have been observed to be
108: the direct progenitors of Type IIP supernovae (Smartt et
109: al. 2004\nocite{2004Sci...303..499S}), but can also evolve towards the
110: blue via a Yellow Hypergiant phase (Oudmaijer et
111: al. 2008\nocite{2008arXiv0801.2315O}; de Wit et
112: al. 2008a\nocite{2008A&A...480..149D}) to become a Wolf-Rayet star and
113: eventually explode as a supernova (Meynet \& Maeder
114: 2000\nocite{2000A&A...361..101M}).
115: 
116: It is not only the study of the stars themselves that helps us
117: understand the final stages of the evolution of massive stars. 
118: Investigating RSGs' mass-loss and their circumstellar
119: material helps us to understand the origin of the aspherical
120: structures found around supernovae, such as SN 1987A's rings, or 
121: gamma-ray bursts.  It is possible that at least some of
122: these observed asymmetries originate during the RSG phase (e.g. Chita et al. 2008). Indeed, in
123: several instances, the mass-loss during the RSG phase has been found
124: to deviate from spherically symmetric, and can have a complex
125: appearance. Much effort has been directed towards the objects VY CMa
126: and NML Cyg, extremely bright, cool, objects close the empirical
127: Humphreys Davidson limit (Humphreys \& Davidson 1979\nocite{1979ApJ...232..409H}). 
128: High resolution studies have
129: revealed highly anisotropic structures in the wind of VY CMa
130: (e.g. Smith et al. 2001\nocite{2001AJ....121.1111S}), while NML Cyg's
131: aspherical appearance is shaped by the strong radiation from the
132: nearby Cyg OB2 association (Morris and Jura 1983; Schuster et al
133: 2006\nocite{2006AJ....131..603S}).
134: 
135: A third RSG that is found in the same location in the HR diagram as
136: the above two objects is $\mu$ Cep (Schuster et al 2006\nocite{2006AJ....131..603S}). 
137: %also known as Herschel's garnet star because of its deep red-orange colour. 
138: Contrary
139: to these two cooler objects, little is known about the material
140: surrounding $\mu$ Cep. This can be readily explained by its mass-loss,
141: which is orders of magnitude lower than for NML Cyg or VY CMa
142: (cf. Jura \& Kleinman 1990\nocite{1990ApJS...73..769J}).  
143: %This M2Ia supergiant has an
144: %interferometrically measured photospheric diameter of 20 milli-arcsec
145: %in the optical (Mozurkewich et al. 2003\nocite{2003AJ....126.2502M}), and is therefore one of the
146: %largest stars on the sky as seen from Earth. 
147: With an $M_{\rm bol}$ of $-$9.08, this M2Ia star is one of the brightest RSGs known, lying on
148: evolutionary tracks corresponding to stars with initial masses greater
149: than 25 M$_{\odot}$ (Levesque et al. 2005\nocite{2005ApJ...628..973L}).  Using Levesque et al.'s
150: parameters, we find a distance to the object of 870 pc, and a
151: luminosity of $\sim$ 3$\times 10^5$ L$_{\odot}$.
152: Here, we report on the discovery of an extended dust shell around the object by means
153: of diffraction limited imaging (0.6\arcsec) of mid-IR thermal dust emission. 
154: 
155: \section{Observations and data reduction}
156: \label{obser}
157: \begin{figure}
158:   %\center{\includegraphics[height=8cm,width=8cm,angle=0]{f1.ps}}
159:   \epsscale{1.0}
160:   \plotone{f1.eps}
161:   \caption[]{The azimuthally averaged and normalized intensity profile of $\mu$ Cep as
162:   observed at 6 different dates. The profile is stable. Note the slightly narrower profile for the 030713 data. These are  
163:   less sensitive, and with a SNR a factor of 10 less
164:   than the other data, the faint outer parts can not be measured. Also shown are the intensity profiles of the 3 genuine PSF objects. Note the first
165: two Airy rings at $\sim1.2\arcsec$ and $\sim2.4\arcsec$.} 
166:   \label{obspsf} 
167: \end{figure}
168: 
169: $\mu$ Cep was observed on 6 different occasions as a mid-IR standard
170: star between June 2003 and July 2004 with the COMICS instrument
171: mounted on the 8.2 meter Subaru telescope in Hawaii (Kataza et
172: al. 2000\nocite{2000SPIE.4008.1144K}; Okamoto et
173: al. 2003\nocite{2003SPIE.4841..169O}; Sako et
174: al. 2003\nocite{2003SPIE.4841.1211S}). The star was imaged using the
175: Q24.5-OLD and Q24.5-NEW filters both centred on 24.5\,\micron. The imaging mode of
176: COMICS utilises a Raytheon 320$\times$240 Si:As IBC array with a pixel
177: scale of 0.13\arcsec. This scale comfortably samples the diffraction
178: limit for the 8.2m Subaru telescope at this wavelength, which is
179: 0.6\arcsec. The observations are summarised
180: in table\,1.  More details on the filters and data reduction can be
181: found in de Wit et al. (2008b, in prep.).
182: 
183: $\mu$ Cep is bright and therefore often used as a standard star,
184: either for determining the instrumental point-spread-function (PSF) or
185: for calibrating photometry. We observed the object for calibration
186: purposes for a different project (as reported in de Wit et al. 2008b),
187: and found that it was extended compared to other PSF reference
188: stars. As $\mu$ Cep was the PSF standard, we do not have a concomitant
189: PSF reference observations. However, we can use as a reference PSF
190: the remaining three genuine PSF objects observed for the same
191: project: $\alpha$ Tau and asteroids 51 and 511. To illustrate the
192: stability of the observational set-up and the extent of $\mu$ Cep, we
193: show in Fig.~\ref{obspsf} the, azimuthally averaged, normalized
194: intensity profiles of the 3 genuine PSF standards and those of each of
195: the 6 observations of $\mu$ Cep.  The 3 point sources have virtually
196: identical profiles, and the noise begins to dominate only after $\sim$4\arcsec.
197: On the contrary, the profiles for $\mu$ Cep are clearly extended with respect
198: to the 3 PSF standards to at least 6\arcsec, and they are also
199: hardly different from each other on all occasions. The lowest signal-to-noise ratio (SNR) data
200: of $\mu$ Cep (taken in July 2003) also reveal that the object is
201: extended, but the data are not deep enough to probe the extent as well
202: as the higher SNR data.  We conclude that $\mu$ Cep is extended at
203: 24.5\,$\mu$m.
204: 
205: \section{Results}
206: \begin{figure*}
207: %  \center{
208: %\includegraphics[height=8cm,width=8cm]{mucep5.ps}
209: %\includegraphics[height=8cm,width=8cm]{Input/mucep1_fil.eps}
210: %\includegraphics[height=8cm,width=8cm]{AlfTau.ps}
211: %\includegraphics[height=8cm,width=8cm]{Input/psf1_fil.eps}
212: \plottwo{f2a.eps}{f2b.eps}
213: %}
214: \caption[]{The PSF subtracted image of $\mu$ Cep (left) and the PSF standard
215:   $\alpha$ Tau (right). The instrumental PSF was determined from the point
216:   source asteroid\,511. The region within 1.2\arcsec~is masked to hide
217:   artefacts from the subtraction procedure. $\mu$ Cep is clearly extended, and
218:   an asymmetry originates from close to the star. The contour levels are
219:   chosen to enhance the contrast between the intensity levels, lowest contour corresponds to twice 
220:   the standard deviation in the background level. North is to the left, East is to the bottom.}
221: \label{images}
222: \end{figure*}
223: 
224: 
225: \subsection{The 24.5\,$\mu$m images: asymmetric dust distribution}
226: 
227: The highest SNR image (taken on 04/06/07) of $\mu$ Cep is shown in the
228: left panel of Fig.~\ref{images}. We have subtracted from this image
229: the scaled profile of the PSF standard asteroid 511 in order to
230: enhance the features of $\mu$ Cep's circumstellar emission. For
231: comparison we show in the right panel of Fig.~\ref{images} the PSF standard 
232: $\alpha$ Tau with again a PSF profile subtracted. It is clear that $\mu$ Cep has extended circumstellar
233: emission reaching distances from the star of at least 6\arcsec. The
234: outer parts of the shell-like structure are roughly circular
235: with indications of larger scale inhomogeneities. In contrast, the inner
236: parts reveal a clearly asymmetric, bipolar type, geometry at a
237: position angle of $\sim$67$^{\rm o}$.  The elliptical structure is
238: visible from the inner parts and extends to 2.2\arcsec~along the major
239: axis on either side of the central star, and 1.4\arcsec~along the
240: minor axis.  The intensity contrast between the brighter regions and
241: the fainter regions at the same distance from the star is a factor of
242: 2.
243: 
244: \subsection{Dust radiative transfer modelling}
245: \label{sec:mod}
246: Emission at 24.5\,\micron~stays optically thin for large columns of dust. 
247: Resolved images allow us to simultaneously model the dust emission as a function
248: of distance from the star (the intensity profile) and the total dust emission
249: as given by the SED. For this purpose, we employ DUSTY, a code that
250: solves self-consistently the scaled 1D dust radiative transfer problem (see
251: Ivezi\'c \& Elitzur 1997\nocite{1997MNRAS.287..799I}). We use a spherically symmetric dust distribution,
252: that is illuminated by a central, unresolved star. Its radiation is
253: represented by a Kurucz ATLAS9 atmosphere model (Kurucz 1993\nocite{1993yCat.6039....0K}) with
254: $T_{\rm eff}=3750\,{\rm K}$, solar metallicity and surface gravity
255: corresponding to the supergiant nature of $\mu$ Cep, i.e. $\rm log(g)=0.0$. 
256: We use oxygen-rich dust with a condensation temperature of 1000\,{\rm K}. The dust particles
257: follow a MRN size distribution (Mathis, Rumple, \& Nordsieck
258: 1977\nocite{1977ApJ...217..425M}). Within the limitation of a spherical model,
259: we experiment with (1) the amount of dust as parametrized by the total optical
260: depth $A_{V}$; (2) the radial density distribution of the
261: dust $\propto r^{-p}$, adopting a constant ($p=2$) or declining ($p=1.5$) mass-loss rate; (3) two different silicate opacity tables, the ones provided by
262: Draine \& Lee (1984\nocite{1984ApJ...285...89D}) and for ``warm'' silicates by 
263: Ossenkopf et al. (1992\nocite{1992A&A...261..567O}). The
264: outer bound of the model is set at 1000 times the dust condensation radius,
265: but the exact value is of little influence on the total excess flux at
266: wavelengths shortward of $\sim$25\,\micron~as long as the outer radius is much larger
267: than the inner dust condensation radius.
268: 
269: We build $\mu$ Cep's SED using a mid-IR spectrum taken with the short
270: wavelength spectrometer (SWS, de Graauw et
271: al. 1996\nocite{1996A&A...315L..49D}) on board the ISO satellite
272: (Kessler et al. 1996\nocite{1996A&A...315L..27K}), near-IR $JHKLM$
273: photometry from Heske (1990)\nocite{1990A&A...229..494H} and visual
274: broadband $UBVRI$ Johnson photometry from Lee (1970\nocite{1970PASP...82..765L}).  The photometry is dereddened
275: for an interstellar extinction of $A_{V}=1.5^{m}$ 
276: %a value based onthe total extinction of $A_{V}=2.0^{m}$ 
277: (Levesque et al. 2005). The IRAS 60\,$\mu$m and 100\,$\mu$m data
278: points lie above the extrapolated ISO-SWS spectrum. This excess flux
279: is due to the much larger beam of IRAS and comes from extended cool
280: dust emission. The origin of this emission is not obvious, it could be
281: either due to the interstellar medium being heated up by the
282: stellar radiation (cf. Oudmaijer 1996), or a very extended shell which
283: is the result of a previous mass-loss phase (e.g. Stencel et
284: al. 1988). It should be kept in mind that we only consider the most
285: recent mass-loss episode here. 
286: 
287: The fit procedure initially estimates the $L_{\rm
288: bol}$ by matching the overall shape of the scaled model SED to the
289: observed one. DUSTY's output images are then accordingly scaled and
290: convolved with the instrumental PSF. A comparison of model infrared
291: excess and model intensity profile to the observed ones is made for
292: all generated models. A simple tally is performed based on a goodness-of-fit 
293: criterion. The model that fits both sets of data best is the
294: one with the highest average ranking in the two tallies.
295: 
296: The results of the 1D modelling are presented in Fig.\,\ref{SED}.
297: %Spherical dust models can provide reasonable though not perfect fits
298: %to the observations. 
299: Once the intensity profile is fit for a given
300: nebular structure and stellar luminosity, the SED can be matched by
301: increasing the amount of dust in the given nebula. The normalized
302: intensity profile is quite insensitive to the total amount of dust for
303: small optical depths. We chose to prioritize a fit to the slope of the
304: continuum longward of the silicate feature rather than the silicate
305: emission profile. Details for each of the three models presented
306: in Fig.\,\ref{SED} are given in Table\,1.  We present the two
307: best-fitting $p=2$ models corresponding to the two different silicate
308: opacities. The silicate emission feature is not well matched by
309: either model, although the intensity profile is better fit by model
310: \#1 (full line). At the inner 2\arcsec~the models predict too much intensity,
311: which is consistent with a deviation from spherical symmetry seen in
312: the images.  Models with a shallow $p=1.5$ radial density profile fit
313: the intensity profile worst as is illustrated by model \#3. This
314: demonstrates the value of spatial information in this type of analysis,
315: because the SED is at the same time very well matched by model
316: \#3. All models have a mass-loss rate of a few $10^{-7}\,\Msunyr$ for an
317: expansion velocity of $\rm 10\,km\,s^{-1}$ (cf. Le Borgne \& Mauron
318: 1989\nocite{1989A&A...210..198L}).
319: 
320: \section{Discussion}
321: We have presented the first high-resolution (0.6\arcsec) images of the
322: circumstellar environment of the RSG $\mu$ Cep. This material has not been seen
323: in previous imaging campaigns. For a sample of massive evolved objects,
324: Schuster et al. (2006) obtained deep optical images with the HST to
325: search for scattered light by circumstellar dust. They detected only those
326: sources which have mass-loss rates many orders of magnitude larger than $\mu$ Cep over the past 500--1000
327: years. Le Borgne \& Mauron (1989) searched for optical, reflected
328: emission in their polarization data. These authors made pointed
329: observations with a 10\arcsec~diaphragm at several locations around
330: the star and detected (faint) polarized light as far as 20\arcsec~from
331: the star. Other evidence for extended emission comes from Mauron (1997\nocite{1997A&A...326..300M}), who finds
332: resonance scattered K{ \sc i} line emission at distances further than
333: 20\arcsec~from the star. These data were taken at 2 slit positions,
334: and thus do not allow for conclusive statements about the geometry of
335: the circumstellar matter to be made. They do appear to indicate the
336: presence of either clumpy material or several discrete mass-loss
337: episodes in the recent past.  The data presented in this paper
338: traces material out to 6\arcsec. As an expansion velocity of
339: 10\,km\,s$^{-1}$ at a distance of 1 kpc corresponds to about 1\arcsec~per
340: 500 yr, the mass lost 2000--3000 years ago appears to have been ejected 
341: in a roughly spherically symmetric way. However, the most recent mass lost, ejected less than
342: 1000 years ago, shows a pronounced axisymmetric geometry in our high resolution
343: images.
344: 
345: \begin{table}
346:    {
347:      \begin{center}
348:        \caption[]{Input parameters and derived quantities of the models presented in Fig\,\ref{SED}. DL and OW dust types stand
349: for Draine \& Lee (1984), and Ossenkopf et al. (1992) silicates. Model \#1 is the preferred model.}
350:        \begin{tabular}{ccccclccl}
351:          \hline
352:          \hline
353:          \#    &  p  & $R_{\rm in}$ & $R_{\rm out}/R_{\rm in}$      & $L$         &   dust & $A_{V}$ &   $M_{\rm dust}$ & $\Mdot$  \\
354:                   &     & (mas)     &    & ($L_{\odot}$)&   &        &  ($ M_{\odot}$)& (\Msunyr)\\
355:  	\hline            
356:         1        &  2.0   & 75   & $1\,10^{3}$  & $3.3\,10^{5}$ & DL  & 0.70  &  $1.4\,10^{-2}$  &  $4.5\,10^{-7}$\\
357:         2        &  2.0   & 103  & $1\,10^{3}$  & $3.1\,10^{5}$ & OW  & 0.60  &  $2.2\,10^{-2}$  &  $5.2\,10^{-7}$\\    
358:         3        &  1.5   & 112  & $1\,10^{2}$  & $3.9\,10^{5}$ & OW  & 0.30  &  $5.0\,10^{-3}$  &  $1.6\,10^{-7}$\\
359:  	\hline
360:        \end{tabular}
361:      \end{center}
362:    }
363:    \label{tab:parameters}
364: \end{table}
365: 
366: 
367: \begin{figure*}
368:   \center{\includegraphics[height=8cm,width=6cm,angle=90]{f3a.eps}
369:     \includegraphics[height=8cm,width=6cm,angle=90]{f3b.eps}}
370: %  \plottwo{Input/muCep_5_SED.ps}{Input/muCep_5_prof.ps}
371:   \caption[]{Model fits to the SED (left panel) and the 24.5\,\micron~intensity profile (right panel). Three models are
372:     shown corresponding to the parameters given in Table\,2. Filled circles represent the extinction corrected optical and near-IR 
373: data (open triangles). Sources of the SED data are discussed in the text. }
374:   \label{SED}
375: \end{figure*}
376: 
377: One may speculate whether this geometry is due to a bipolar flow colliding with the
378: spherically symmetric previous wind, or a slowly expanding torus.
379: It is useful in this respect to consider the current findings for the
380: more numerous lower-mass counterparts of RSGs, the
381: Asymptotic Giant Branch (AGB) stars. Data on these are also sparse,
382: but asymmetries of the circumstellar material have been seen at high-resolution 
383: in several cases (e.g. Menut et al. 2007\nocite{2007MNRAS.376L...6M} and Murakawa et
384: al. 2005\nocite{2005A&A...436..601M} for IRC +10216; 
385: Vinkovi\'c et al. 2004\nocite{2004MNRAS.352..852V} and Inomata et
386: al. 2007\nocite{2007PASJ...59..799I} 
387: for IRC +10011). The picture that emerges from such individual studies
388: is that the outer parts of the AGB stars are fairly spherical. The inner parts are 
389: relatively complex, and can be best  explained with a torus type structure 
390: close to the star, perhaps carved out by a bi-polar flow. 
391: %IRC +10011 is a good example of this,
392: %while IRC +10216 could be further in its evolution (cf. Vinkovi\'c et
393: %al. 2004).
394: Based on a compilation of the best data available, Huggins (2007\nocite{2007ApJ...663..342H})
395: finds that, in general, tori correspond to low outflow velocities and
396: jets with higher speeds, in excess of 100\,km\,s$^{-1}$. 
397: %The shaping of
398: %the circumstellar shells around AGB stars occurs during the later
399: %stages of the AGB phase.  
400: In addition, he concludes that the jets
401: develop shortly after a torus is ejected. Given that the CO spectra of
402: $\mu$ Cep indicate low velocities rather than high velocities, it is
403: very well possible that we are now witnessing the same for a Red
404: Supergiant.
405: 
406: \section{Concluding Remarks}
407: 
408: We have obtained the first diffraction-limited images of a Red
409: Supergiant at 24.5\,$\mu$m, and resolved the circumstellar material
410: around $\mu$ Cep out to 6\arcsec. The intensity profile and the SED 
411: were simultaneously fitted with a dust model. The main results can be summarized as follows:
412: 
413: %(MORE  : DEPENDENT ON SED FITTING RESULTS)
414: 
415: 1. The outer parts of the shell are to first order circular with apparent inhomogeneities, the
416:   inner parts, tracing the mass lost in the past 1000 years display a
417:   flattened structure, which is possibly a slowly expanding, dense
418:   torus. The immediate conclusion from this is that any asymmetries in
419:   the shells of massive evolved stars and Supernovae ejecta may find their
420:   origins in the Red Supergiant phase. This is qualitatively very
421:   similar to what is found for the lower-mass AGB stars which evolve
422: into bipolar Planetary Nebulae.
423: 
424: 2. The mass-loss rate for $\mu$ Cep is found to be a few times 10$^{-7}$\,M$_{\odot}\,$yr$^{-1}$. This low mass-loss rate may explain the fact
425:   that $\mu$ Cep is not detected in scattered light, as opposed to other
426:   RSGs which have mass-loss rates that are orders of magnitude larger.
427:   
428: 3. A general conclusion is that imaging thermal dust emission at
429:   24.5\,$\mu$m is a viable manner to spatially resolve the
430:   circumstellar material around evolved objects with low
431:   mass-loss rates and as a consequence are otherwise not detected with
432:   maser observations, CO imaging or scattered light imaging.  
433: %Here we
434: %  have shown that diffraction-limited thermal imaging can readily
435: %  detect and resolve such structures.  
436: 
437: %(ALSO MENTION OHNAKA BIPOLAR VS TORUS)
438: %MASER SPOTS UNCLEAR NML CYG UNCLEAR
439: 
440: 
441: %\begin{acknowledgements}
442: \acknowledgements
443: RDO is grateful for the support from the Leverhulme Trust for awarding
444: a Research Fellowship. We thank Martin Groenewegen for fruitful discussions.
445: The version of the ISO data presented in this paper correspond to the 
446: Highly Processed Data Product (HPDP) set called hpdp\_39802402\_5 
447: by W.F. Frieswijk et al., available for public use in the ISO Data Archive.
448: %\end{acknowledgements}
449: 
450: \bibliographystyle{aa}
451: %\bibliography{mucep.bib}
452: \begin{thebibliography}{32}
453: \expandafter\ifx\csname natexlab\endcsname\relax\def\natexlab#1{#1}\fi
454: 
455: \bibitem[{{de Graauw} {et~al.}(1996){de Graauw}, {Haser}, {Beintema},
456:   {Roelfsema}, {van Agthoven}, {Barl}, {Bauer}, {Bekenkamp}, {Boonstra},
457:   {Boxhoorn}, {Cote}, {de Groene}, {van Dijkhuizen}, {Drapatz}, {Evers},
458:   {Feuchtgruber}, {Frericks}, {Genzel}, {Haerendel}, {Heras}, {van der Hucht},
459:   {van der Hulst}, {Huygen}, {Jacobs}, {Jakob}, {Kamperman}, {Katterloher},
460:   {Kester}, {Kunze}, {Kussendrager}, {Lahuis}, {Lamers}, {Leech}, {van der
461:   Lei}, {van der Linden}, {Luinge}, {Lutz}, {Melzner}, {Morris}, {van Nguyen},
462:   {Ploeger}, {Price}, {Salama}, {Schaeidt}, {Sijm}, {Smoorenburg}, {Spakman},
463:   {Spoon}, {Steinmayer}, {Stoecker}, {Valentijn}, {Vandenbussche}, {Visser},
464:   {Waelkens}, {Waters}, {Wensink}, {Wesselius}, {Wiezorrek}, {Wieprecht},
465:   {Wijnbergen}, {Wildeman}, \& {Young}}]{1996A&A...315L..49D}
466: {de Graauw}, T., {Haser}, L.~N., {Beintema}, D.~A., {et~al.} 1996, \aap, 315,
467:   L49
468: 
469: \bibitem[]{2008arxiv} {Chita}, S.~M. and {Langer}, N. and {van Marle}, A.~J. and {Garc{\'{\i}}a-Segura}, G. and 
470: 	{Heger}, A., 2008, ArXiV 0807.3049
471: 
472: \bibitem[{{de Wit} {et~al.}(2008){de Wit}, {Oudmaijer}, {Groenewegen}, {Hoare},
473:   \& {Malbet}}]{2008A&A...480..149D}
474: {de Wit}, W.~J., {Oudmaijer}, R.~D., {Groenewegen}, M.~A.~T., {Hoare}, M.~G.,
475:   \& {Malbet}, F. 2008a, \aap, 480, 149
476: 
477: \bibitem[]{2008subm} {de Wit}, W.~J., {Hoare}, M.~G., {Fujiyoshi}, T., {Oudmaijer}, R.~D., {Honda}, M., {Kataza}, H., {Miyata}, T., {K. Okamoto}, Y., {Onaka}, T., {Sako}, S., {Yamashita}, T. 2008b, \aap, subm.
478: 
479: 
480: \bibitem[{{Draine} \& {Lee}(1984)}]{1984ApJ...285...89D}
481: {Draine}, B.~T. \& {Lee}, H.~M. 1984, \apj, 285, 89
482: 
483: \bibitem[{{Heske}(1990)}]{1990A&A...229..494H}
484: {Heske}, A. 1990, \aap, 229, 494
485: 
486: \bibitem[{{Huggins}(2007)}]{2007ApJ...663..342H}
487: {Huggins}, P.~J. 2007, \apj, 663, 342
488: 
489: \bibitem[{{Humphreys} \& {Davidson}(1979)}]{1979ApJ...232..409H}
490: {Humphreys}, R.~M. \& {Davidson}, K. 1979, \apj, 232, 409
491: 
492: \bibitem[{{Inomata} {et~al.}(2007){Inomata}, {Imai}, \&
493:   {Omodaka}}]{2007PASJ...59..799I}
494: {Inomata}, N., {Imai}, H., \& {Omodaka}, T. 2007, \pasj, 59, 799
495: 
496: \bibitem[{{Ivezic} \& {Elitzur}(1997)}]{1997MNRAS.287..799I}
497: {Ivezi\'c}, Z. \& {Elitzur}, M. 1997, \mnras, 287, 799
498: 
499: 
500: \bibitem[{{Jura} \& {Kleinmann}(1990)}]{1990ApJS...73..769J}
501: {Jura}, M. \& {Kleinmann}, S.~G. 1990, \apjs, 73, 769
502: 
503: %\bibitem[{{Kataza} {et~al.}(2000){Kataza}, {Okamoto}, {Takubo}, {Onaka},
504: %  {Sako}, {Nakamura}, {Miyata}, \& {Yamashita}}]{2000SPIE.4008.1144K}
505: %{Kataza}, H., {Okamoto}, Y., {Takubo}, S., {et~al.} 2000, in SPIE, Vol. 4008,
506: %  {Optical and IR Telescope Instrumentation and Detectors}, ed. M.~{Iye} \&
507: %  A.~F. {Moorwood}, 1144
508: 
509: \bibitem[{{Kataza} {et~al.}(2000){Kataza}, {Okamoto}, {Takubo}, {Onaka},
510:   {Sako}, {Nakamura}, {Miyata}, \& {Yamashita}}]{2000SPIE.4008.1144K}
511: {Kataza}, H., {Okamoto}, Y., {Takubo}, S., {et~al.} 2000, in SPIE, Vol. 4008, eds. M.~{Iye} \&  A.~F. {Moorwood}, 1144
512: 
513: \bibitem[{{Kessler} {et~al.}(1996){Kessler}, {Steinz}, {Anderegg}, {Clavel},
514:   {Drechsel}, {Estaria}, {Faelker}, {Riedinger}, {Robson}, {Taylor}, \&
515:   {Xim{\'e}nez de Ferr{\'a}n}}]{1996A&A...315L..27K}
516: {Kessler}, M.~F., {Steinz}, J.~A., {Anderegg}, M.~E., {et~al.} 1996, \aap, 315,
517:   L27
518: 
519: \bibitem[{{Kurucz}(1993)}]{1993yCat.6039....0K}
520: {Kurucz}, R.~L. 1993, VizieR Online Data Catalog, 6039
521: 
522: \bibitem[{{Le Borgne} \& {Mauron}(1989)}]{1989A&A...210..198L}
523: {Le Borgne}, J.~F. \& {Mauron}, N. 1989, \aap, 210, 198
524: 
525: \bibitem[{{Lee}(1970)}]{1970PASP...82..765L}
526: {Lee}, T.~A. 1970, \pasp, 82, 765
527: 
528: \bibitem[{{Levesque} {et~al.}(2005){Levesque}, {Massey}, {Olsen}, {Plez},
529:   {Josselin}, {Maeder}, \& {Meynet}}]{2005ApJ...628..973L}
530: {Levesque}, E.~M., {Massey}, P., {Olsen}, K.~A.~G., {et~al.} 2005, \apj, 628,
531:   973
532: 
533: \bibitem[{{Massey} {et~al.}(2008){Massey}, {Levesque}, {Plez}, \&
534:   {Olsen}}]{2008arXiv0801.1806M}
535: {Massey}, P., {Levesque}, E.~M., {Plez}, B., \& {Olsen}, K.~A.~G. 2008, ArXiv
536:   e-prints, 0801.1806
537: 
538: \bibitem[{{Mathis} {et~al.}(1977){Mathis}, {Rumpl}, \&
539:   {Nordsieck}}]{1977ApJ...217..425M}
540: {Mathis}, J.~S., {Rumpl}, W., \& {Nordsieck}, K.~H. 1977, \apj, 217, 425
541: 
542: \bibitem[{{Mauron}(1997)}]{1997A&A...326..300M}
543: {Mauron}, N. 1997, \aap, 326, 300
544: 
545: \bibitem[{{Menut} {et~al.}(2007){Menut}, {Gendron}, {Schartmann}, {Tuthill},
546:   {Lopez}, {Danchi}, {Wolf}, {Lagrange}, {Flament}, {Rouan}, {Cl{\'e}net}, \&
547:   {Berruyer}}]{2007MNRAS.376L...6M}
548: {Menut}, J.-L., {Gendron}, E., {Schartmann}, M., {et~al.} 2007, \mnras, 376, L6
549: 
550: \bibitem[{{Meynet} \& {Maeder}(2000)}]{2000A&A...361..101M}
551: {Meynet}, G. \& {Maeder}, A. 2000, \aap, 361, 101
552: 
553: \bibitem[]{1983ApJ...267..179M}{{Morris}, M. and {Jura}, M.}, 1983, ApJ 267, 179
554: 
555: %\bibitem[{{Mozurkewich} {et~al.}(2003){Mozurkewich}, {Armstrong}, {Hindsley},
556: %  {Quirrenbach}, {Hummel}, {Hutter}, {Johnston}, {Hajian}, {Elias}, {Buscher},
557: %  \& {Simon}}]{2003AJ....126.2502M}
558: %{Mozurkewich}, D., {Armstrong}, J.~T., {Hindsley}, R.~B., {et~al.} 2003, \aj,
559: %  126, 2502
560: 
561: \bibitem[{{Murakawa} {et~al.}(2005){Murakawa}, {Suto}, {Oya}, {Yates}, {Ueta},
562:   \& {Meixner}}]{2005A&A...436..601M}
563: {Murakawa}, K., {Suto}, H., {Oya}, S., {et~al.} 2005, \aap, 436, 601
564: 
565: %\bibitem[{{Okamoto} {et~al.}(2003){Okamoto}, {Kataza}, {Yamashita}, {Miyata},
566: %  {Sako}, {Takubo}, {Honda}, \& {Onaka}}]{2003SPIE.4841..169O}
567: %{Okamoto}, Y.~K., {Kataza}, H., {Yamashita}, T., {et~al.} 2003, in SPIE, Vol.
568: %  4841, Instrument Design and Performance for Optical/Infrared Ground-based
569: %  Telescopes, ed. M.~{Iye} \& A.~F.~M. {Moorwood}, 169
570: 
571: \bibitem[{{Okamoto} {et~al.}(2003){Okamoto}, {Kataza}, {Yamashita}, {Miyata},
572:   {Sako}, {Takubo}, {Honda}, \& {Onaka}}]{2003SPIE.4841..169O}
573: {Okamoto}, Y.~K., {Kataza}, H., {Yamashita}, T., {et~al.} 2003, in SPIE, Vol.
574:   4841, eds. M.~{Iye} \& A.~F.~M. {Moorwood}, 169
575: 
576: \bibitem[{{Ossenkopf} {et~al.}(1992){Ossenkopf}, {Henning}, \&
577:   {Mathis}}]{1992A&A...261..567O}
578: {Ossenkopf}, V., {Henning}, T., \& {Mathis}, J.~S. 1992, \aap, 261, 567
579: 
580: \bibitem[]{704} Oudmaijer R.D. 1996, A\&A 306, 823
581: 
582: \bibitem[{{Oudmaijer} {et~al.}(2008){Oudmaijer}, {Davies}, {de Wit}, \&
583:   {Patel}}]{2008arXiv0801.2315O}
584: {Oudmaijer}, R.~D., {Davies}, B., {de Wit}, W.-J., \& {Patel}, M. 2008, ArXiv
585:   e-prints, 0801.2315
586: 
587: %\bibitem[{{Sako} {et~al.}(2003){Sako}, {Kataza}, {Miyata}, {Okamoto}, {Takubo},
588: %  {Honda}, {Onaka}, \& {Yamashita}}]{2003SPIE.4841.1211S}
589: %{Sako}, S., {Kataza}, H., {Miyata}, T., {et~al.} 2003, in SPIE, Vol. 4841,
590: %  Instrument Design and Performance for Optical/Infrared Ground-based
591: %  Telescopes, ed. M.~{Iye} \& A.~F.~M. {Moorwood}, 1211
592: \bibitem[{{Sako} {et~al.}(2003){Sako}, {Kataza}, {Miyata}, {Okamoto}, {Takubo},
593:   {Honda}, {Onaka}, \& {Yamashita}}]{2003SPIE.4841.1211S}
594: {Sako}, S., {Kataza}, H., {Miyata}, T., {et~al.} 2003, in SPIE, Vol. 4841, eds. M.~{Iye} \& A.~F.~M. {Moorwood}, 1211
595: 
596: \bibitem[{{Schuster} {et~al.}(2006){Schuster}, {Humphreys}, \&
597:   {Marengo}}]{2006AJ....131..603S}
598: {Schuster}, M.~T., {Humphreys}, R.~M., \& {Marengo}, M. 2006, \aj, 131, 603
599: 
600: \bibitem[{{Smartt} {et~al.}(2004){Smartt}, {Maund}, {Hendry}, {Tout},
601:   {Gilmore}, {Mattila}, \& {Benn}}]{2004Sci...303..499S}
602: {Smartt}, S.~J., {Maund}, J.~R., {Hendry}, M.~A., {et~al.} 2004, Science, 303,
603:   499
604: 
605: \bibitem[{{Smith} {et~al.}(2001){Smith}, {Humphreys}, {Davidson}, {Gehrz},
606:   {Schuster}, \& {Krautter}}]{2001AJ....121.1111S}
607: {Smith}, N., {Humphreys}, R.~M., {Davidson}, K., {et~al.} 2001, \aj, 121, 1111
608: 
609: %\bibitem[{{Tody}(1993)}]{1993ASPC...52..173T}
610: %{Tody}, D. 1993, in ASPC, Vol.~52, Astronomical Data Analysis Software and
611: %  Systems II, ed. R.~J. {Hanisch}, R.~J.~V. {Brissenden}, \& J.~{Barnes}, 173
612: 
613: \bibitem[]{1988AJ.....95..141S}{Stencel}, R.~E. and {Pesce}, J.~E. and {Hagen Bauer}, W. 1988, AJ,95, 141
614: 
615: \bibitem[{{van Loon} {et~al.}(2005){van Loon}, {Cioni}, {Zijlstra}, \&
616:   {Loup}}]{2005A&A...438..273V}
617: {van Loon}, J.~T., {Cioni}, M.-R.~L., {Zijlstra}, A.~A., \& {Loup}, C. 2005,
618:   \aap, 438, 273
619: 
620: \bibitem[{{Vinkovi{\'c}} {et~al.}(2004){Vinkovi{\'c}}, {Bl{\"o}cker},
621:   {Hofmann}, {Elitzur}, \& {Weigelt}}]{2004MNRAS.352..852V}
622: {Vinkovi{\'c}}, D., {Bl{\"o}cker}, T., {Hofmann}, K.-H., {Elitzur}, M., \&
623:   {Weigelt}, G. 2004, \mnras, 352, 852
624: 
625: \end{thebibliography}
626: 
627: 
628: 
629: 
630: 
631: 
632: 
633: \end{document}
634: