0709.0488/ms.tex
1: 
2: 
3: %% manuscript produces a one-column, double-spaced document:
4: 
5: \documentclass[12pt,preprint]{aastex}
6: 
7: %% preprint2 produces a double-column, single-spaced document:
8: 
9: %
10: \newcommand\ms{M$_{\odot}$}
11: \newcommand\lum{$\log L/L_{\sun}$}
12: \newcommand\tef{$\log T_{\rm eff}$}
13: \newcommand\hb{H${\beta}$~}
14: \newcommand\ha{H${\alpha}$~}
15: 
16: 
17: 
18: \slugcomment{The Astrophysical Journal, submitted}
19: 
20: \shorttitle{Dust in Magellanic Cloud Planetary Nebulae}
21: \shortauthors{Stanghellini et~al.}
22:   
23: \begin{document}
24:   
25: \title{{\it Spitzer} Infrared Spectrograph Observations of Magellanic Cloud Planetary Nebulae: 
26: the nature of dust in low metallicity circumstellar ejecta\altaffilmark{1}}
27: 
28: \author{Letizia Stanghellini}
29: \affil{National Optical Astronomy Observatory, 950 N. Cherry Av.,
30: Tucson, AZ 85719}
31: \email{lstanghellini@noao.edu}
32: 
33: \author{Pedro Garc\'\i a-Lario}
34: \affil{Herschel Science Centre, European Space Astronomy Centre, Research and Scientific Support Department of ESA, 
35: Villafranca del Castillo, P.O. Box 50727. E-28080 Madrid, Spain}
36: \email{Pedro.Garcia-Lario@sciops.esa.int}
37: 
38: 
39: \author{D.~Anibal ~Garc\'\i a-Hern\'andez}
40: \affil{The W. J. 
41: McDonald Observatory. The University of Texas at Austin, 1 University Station C1402, Austin, TX 78712}
42: \email{agarcia@astro.as.utexas.edu}
43: 
44: \author{Jose V.~Perea-Calder\'on}
45: \affil{European Space Astronomy Centre, INSA S.~A., P.O. Box 50727. E-28080 Madrid, Spain}
46: \email{Jose.Perea@sciops.esa.int}
47: 
48: \author{James E. Davies}
49: \affil{National Optical Astronomy Observatory, 950 N. Cherry Av.,
50: Tucson, AZ 85719}
51: \email{jdavies@noao.edu}
52: 
53: 
54: \author{Arturo Manchado}
55: \affil{Instituto de Astrof\'isica de Canarias, v\'{\i}a L\'actea s/n, La Laguna, E-38200 Tenerife, 
56: Spain; affiliated to CSIC, Spain}
57: \email{amt@iac.es}
58: 
59: 
60: \author{Eva Villaver}
61: \affil{Space Telescope Science Institute, 3700 San Martin Drive, Baltimore MD 21218; 
62: affiliated to the Hubble Space Telescope Department of ESA}
63: \email{villaver@stsci.edu}
64: 
65: 
66: \author{Richard A.~Shaw}
67: \affil{National Optical Astronomy Observatory, 950 N. Cherry Av.,
68: Tucson, AZ 85719}
69: \email{shaw@noao.edu}
70: 
71: \altaffiltext{1}{based on observations made with the {\it Spitzer Space Telescope},
72: which is operated by the Jet Propulsion Laboratory, California Institute of Technology, under a contract with NASA.}
73: 
74: 
75: \begin{abstract}
76: We present 5--40 $\mu$m spectroscopy of 41 planetary nebulae (PNe) in the 
77: Magellanic Clouds, observed with the Infrared Spectrograph on board the \textit{Spitzer 
78: Space Telescope}. The spectra show the presence of a combination 
79: of nebular emission lines and solid-state features from dust, superimposed on the thermal IR continuum. 
80: 
81: By analyzing the 25 LMC and 16 SMC PNe in our sample we found that
82: the IR spectra of 14 LMC and 4 SMC PNe
83: are dominated by nebular emission lines, while the other spectra 
84: show solid-state features. We observed that
85: the solid-state features 
86: are compatible with carbon-rich dust grains (SiC, polycyclic aromatic hydrocarbons (PAHs), etc.) in most cases, 
87: except in three PNe showing oxygen-rich dust features. The frequency of carbonaceous dust features is generally higher in
88: LMC than in SMC PNe. 
89: 
90: 
91: The spectral analysis allowed the correlations of the dust characteristics with the
92: gas composition and morphology, and the properties of the central stars. We found that:
93: 1) all PNe with carbonaceous dust features have C/O$>$1, none of these being bipolar or otherwise
94: highly asymmetric;
95: 2) all PNe with oxygen-rich dust features have C/O$<$1, with probable high mass progenitors
96: if derived from single-star evolution (these PNe are either bipolar or highly asymmetric); 
97: 3) the dust temperature tracks the nebular and stellar evolution; and
98: 4) the dust production efficiency depends on metallicity, with low metallicity environments not favoring 
99: dust production.
100: 
101: 
102: 
103: \end{abstract}
104: 
105: \keywords{Planetary nebulae; Magellanic Clouds; Dust; solid state features; morphology}
106: 
107: \section{Introduction}
108: 
109: Planetary nebulae (PNe) are the envelopes ejected toward the end of the evolution 
110: of low- and intermediate-mass stars (1--8 M$_{\odot}$), at the tip of the thermally pulsing 
111: asymptotic giant branch (TP-AGB) phase. The gas and dust ejected in this phase 
112: contain elements that had been produced by nucleosynthesis and then carried to 
113: the stellar surface by convective dredge-up processes. 
114: The slow wind that 
115: eventually produces the PN, also called the superwind, starts with the AGB thermal 
116: pulses and carries most of the envelope mass during the last few thermal pulses 
117: (Vassiliadis \& Wood 1993). The composition of the ejecta is tightly correlated to the
118: initial mass and metallicity of the evolving star: higher masses in this range will have undergone
119: hot bottom burning (HBB) nucleosynthesis, which depletes carbon and 
120: increases nitrogen in PNe (e.g., Marigo 2001).
121: The exact mass-loss mechanism and superwind onset are unclear to date, and, most importantly, 
122: their dependence on the mass and metallicity of the stellar progenitor is unconstrained. 
123: 
124: 
125: Dust is formed on the cool surface of the AGB star, and radiation pressure on the dust grains 
126: may be a dominant factor in the superwind onset. Furthermore, the nature of the
127: dust grains affects dust opacity, thus the role of dust grains in the superwind efficiency 
128: must be related to the chemistry of the nebula. Interestingly, Willson (2000) showed that
129: mass loss might also occur in the absence of dust. 
130: 
131: A PN is born when the circumstellar matter ejected during the superwind phase is 
132: ionized by the UV flux from the central star. 
133: Dust in PNe survives the sputtering due to the UV radiation field for a time period
134: that depends on the nature of the dust grains, rather than on the expansion rate and the 
135: luminosity of the central star (e.g., Kaeufl et al.\ 1993). 
136: Dust in Galactic PNe has been observed extensively. It has been shown that about 40\% of the emergent bolometric 
137: flux in Galactic PNe occurs between 5 and 60 $\mu$m (Zhang \& Kwok 1991) in the form of a blackbody-like 
138: continuum spectrum due to dust scattering (Cohen \& Barlow 1974). Dust in Galactic PNe has been 
139: characterized by its cool (100--200 K) temperature, producing a spectrum peaking at 25--60 $\mu$m.
140: The physical extent of the dust coincides with the gas 
141: emission in most nebulae and is even more extended in others (Volk 2003). 
142: 
143: 
144: Observations of AGB stars and PNe in the Galaxy and the Large (LMC) and Small (SMC) Magellanic Clouds, and
145: other Local Group galaxies provide increasing evidence that metallicity may play a
146: role in the AGB and beyond (i.~e., Garc{\'{\i}}a-Hern{\'a}ndez et al. 2007). 
147: On average, the LMC and SMC metallicities are respectively $\sim$50\% and 25\% that of the Galaxy
148: (Caputo et al.\ 1999).
149: The absence of heavily obscured AGB stars in the Magellanic Clouds (Groenewegen et al.\ 2000; Trams et al.\ 1999), 
150: in contrast with
151: their Galactic counterparts, suggests that, on average, lower metallicity environments
152: are less favorable to dust production. 
153: The ratio of carbon-rich to oxygen-rich AGB stars in Local Group galaxies decreases with increasing metallicity
154: (Cioni \& Habing 2003; Cioni et al. 2003; Schultheis et al. 2004). 
155: AGB stars in low-metallicity galaxies would take a long time to 
156: lose their envelopes if the mass-loss efficiency was proportional to dust content.
157: Their lifetime as AGB stars could be very long, increasing the chance of the
158: nuclear-processed material to surface in their lifetime. In this way, a higher 
159: concentration of nuclear-reaction products is explained.
160: In contrast, high-metal stars would have powerful dust-driven mass-loss rates.
161: As a consequence, their chemistry will be less affected.
162: Stanghellini et al. (2003) analyzed all Magellanic Cloud PNe that have been previously observed with the
163: {\it HST} cameras and found that the frequency of asymmetric (e.~g., bipolar, quadrupolar) PNe
164: in the SMC is only 60$\%$ that
165: of the LMC, where the metallicity is, on average, higher. 
166: One may infer that bipolar
167: morphology is less favorable in the low-metal environment of the SMC than in the 
168: LMC and the Galaxy.
169: Aspherical PNe, especially the bipolar ones, are typically more massive
170: than their spherical counterparts, indicating a more efficient mass-loss mechanism (Peimbert \& Serrano 1980). 
171: Furthermore, the mere bipolar shape indicates the presence of a gas-dust torus around the central star,
172: constraining the outflow into bipolar shape. It is no coincidence that the internal extinction of
173: bipolar PNe is higher than that of spherical PNe. It occurred to us that
174: the different morphologies may 
175: form with completely different mass-loss mechanisms, and that the low-metallicity, low-dust
176: environments may produce only spherical PNe.
177: 
178: From the above described observing results, it is obvious that metallicity affects the 
179: properties of PNe, so that it becomes clear that the dust properties in PNe 
180: need to be studied in a variety of
181: metallicity environments, and there is a good reason to observe dust features in 
182: relation to the absolute physical characteristics of the nebulae and stars. 
183: These are the scientific strands that motivated us to perform the IRS observations on the
184: Magellanic Cloud PNe presented in this paper.
185: While Galactic PNe have been studied with IRAS and ISO 
186: (Zhang \& Kwok 1991, Garc{\'{\i}}a-Lario \& Perea Calder{\' o}n 2003), 
187: most Magellanic Cloud PNe were beyond the reach of these IR telescopes.
188: Magellanic Cloud PNe
189: have known distances, thus lowering the number of free parameters in the 
190: comparison between the data and the models. 
191: 
192: This paper presents low resolution IRS {\it Spitzer} spectra for 25 LMC and 16 SMC 
193: PNe. In \S2 we present the observations relative to this data-set and the data 
194: analysis, including the spectral analysis and a description of 
195: individual nebulae. Section 3 presents the study of the correlations between 
196: the dusty nature of the PNe and their gas morphology and evolution, $\S$4 
197: contains the discussion of our findings, 
198: and $\S$5
199: summarizes our conclusions and suggests future endeavors in this field. 
200: 
201: 
202: 
203: \section{Observations and data analysis}
204: 
205: \subsection{Sample selection}
206: 
207: Our goal is to
208: study the infrared spectral properties of PNe in the Magellanic Clouds, and to relate them
209: to morphology and other nebular  properties, and to the nature of their central stars,
210: in order to gain insight on their formation and evolution.
211: In order to meet our science goals we have selected LMC and SMC PNe
212: that satisfy the following selection: 
213: (1) All targets have been previously observed with the {\it HST} cameras (Shaw et al. 2001, 2006; Stanghellini et al. 1999, 2002), 
214: thus morphological (and, in many cases, stellar) information is available.
215: (2) All targets are smaller than 2\arcsec\ in apparent size, 
216: to facilitate accurate peak-up observations. This requirement excludes only a very small fraction
217: of Magellanic Cloud PNe that have been observed with {\it HST}. These restrictions do not 
218: exclude evolved PNe significantly: at the distance of the Magellanic Clouds, 2\arcsec\ implies a PN diameter between 0.5 pc (for the LMC) and 0.6 pc (for the SMC). Thus, for typical PN expansion velocities our sample  includes PNe up to $\sim$9000 yr after ejection (see models by Villaver et al. 2002), which is a substantial fraction of their visible lifetime. 
219: (3) The abundances of the elements that are related to AGB evolution and population (He, N, O, Ar, S) 
220: are available in the literature. 
221: (4) Carbon abundances have been measured from their UV spectra (Stanghellini et al. 2005, Leisy \& Dennefeld 1996).
222: 
223: After filtering the Magellanic Cloud PNe sample through the above selection, we also excluded those targets already in the Reserved Observation Catalog (ROC) through the {\it Spitzer} GTO or other programs. We shall study those targets in the future through the {\it Spitzer} Data Archive. In Figure 1 we show the histogram of the 
224: [\ion{O}{3}] $\lambda$5007 flux distribution of all Magellanic Cloud PNe that satisfy selection rules (1) and (2)
225: above. These include targets observed by us (black histogram), and those in the ROC (gray histogram). 
226: The sample presented in this paper is most representative 
227: of the [\ion{O}{3}]-bright Magellanic Cloud PNe, where the [\ion{O}{3}] $\lambda$5007 brightness is a good proxy for the 
228: optical PN brightness for all but the most unevolved PNe (Stanghellini et al. 2002). 
229: However, there is increasing evidence for a significant population of very low excitation (VLE) PNe in both the LMC (Reid \& Parker 2006b) and the SMC (Jacoby \& De Marco 2002), where [\ion{O}{3}] is very weak and the 
230: F([\ion{N}{2}])/F(\ha) ratio can be very strong. While it is not clear 
231: that a sample selection based on [\ion{O}{3}] brightness inherently selects 
232: against a particular population of PNe, VLE nebulae may nevertheless be 
233: under-represented in our sample. 
234: Additional mid-IR spectra of VLE nebulae will be needed to understand the 
235: applicability of the results presented here to this class of PNe. In Table 1 we
236: list the selected targets, with their angular radii and morphologies derived
237: from the {\it HST} images (Shaw et al. 2001, 2006; Stanghellini et al. 1999,
238: 2002).
239: 
240: 
241: 
242: 
243: \subsection{Spitzer Observations}
244: 
245: 
246: The spectral data were acquired with the Infrared Spectrograph (IRS, Houck et al.
247: 2004) on board the {\it Spitzer Space Telescope} (Werner et al. 2004) between 2005 July 14 and
248: November 14 as part of a General Observer program ($\#$20443).The observing log is given in Table 2, 
249: where we list the target name, the coordinates from the {\it HST}
250: images, the observing date, the dataset ID number, and the observing campaign. The 5--40 $\mu$m range 
251: spectra were acquired by using the Spitzer 5.2--8.7 $\mu$m Short Low 2nd order (SL2), the
252: 7.4--14.5 $\mu$m Short Low
253: 1st order (SL1), the 14.0--21.3 $\mu$m Long Low 2nd order (LL2), and the 19.5--38.0
254: $\mu$m Long Low 1st order (LL1) IRS modes. Each of the observations consists of 3 cycles of 14 sec
255: and 30 sec duration each for the SL and LL segments respectively.
256: We also performed peak-up on a nearby star to achieve
257: accurate (0.4$''$) pointing. We obtained spectra of 41 PNe. The target SMP~LMC~30, originally on our target list,
258: was not pointed accurately enough for spectral extraction.
259: 
260: The data were first processed by the {\it Spitzer Science Center} (SSC) using the
261: standard data reduction pipeline version 12.4 (except for one source, SMP~LMC~100,
262: which was processed by the pipeline version 13, which was suitable for this dataset). This process includes the 
263:  linearization correction, the subtraction of darks, and cosmic ray
264: removal. The images were also corrected for stray light, and a flat-field correction was applied 
265: to account for pixel-to-pixel response variations. 
266: Through the IRS data reduction pipeline we obtained three Basic Calibrated Data (BCD) files
267: for each slit nod position. These three cycles were then combined by the pipeline
268: using a signal-weighted average to produce the co-added 2-D images. 
269: 
270: The pipeline-processed 2-D images were further cleaned by interpolating over the
271: flagged data and rogue pixels, which depend on the observation campaign. 
272: Sky background was removed by using the image degeneracy relative to
273: the two nod positions. Spectral extraction, and wavelength and flux
274: calibration were performed with the {\it Spitzer IRS Custom Extractor} (SPICE), with point source
275: aperture. The resulting 1-D spectra were cleaned for bad pixels, spurious jumps and
276: glitches, smoothed and merged into one final spectrum per module for each source
277: using the IRS Spectroscopy Modeling Analysis and Reduction Tool 
278: (SMART\footnote{SMART was developed by the
279: IRS Team at Cornell University and is available through the {\it Spitzer Science Center}
280: at Caltech.}). The overlap regions
281: between orders in the four modules were masked by hand to produce a smooth,
282: artifact-free spectrum in the 5.2--38 $\mu$m range. The
283: third {\it bonus} order contributed by the SL and LL modules were also masked, so in
284: the end, each PN has two spectra, one for each nod position.
285: 
286: \subsection{Spectral analysis}
287: 
288: The infrared emission from the target PNe is composed of one or more of the following 
289: components: narrow emission lines from collisionally excited atomic species (nebular 
290: lines), thermal continuum emission, and  
291: emission from dust grains. Most of our IRS spectra show continuum emission at different 
292: brightness levels, and nebular emission lines and solid state features superimposed 
293: on the continuum. We analyzed each of these components in turn. 
294: 
295: We identified and measured the nebular lines in the PN spectra by using the line-fitting 
296: routine in SMART. A single-order fit was made to the baseline for each nebular line, 
297: and the line was then fitted using a single Gaussian. The measured line strengths and 
298: widths will be presented in a future paper, together with a detailed nebular analysis. 
299: We found a spurious feature in the spectrum, nominally at 19 $\mu$m, whose 
300: detection was possible due to its slight offset between nodding positions. In some 
301: cases it falls very close to the [\ion{S}{3}] line at 18.7~$\mu$m, and in these cases the 
302: feature was removed before the measurement of the [\ion{S}{3}] line.
303: From an inventory of the nebular emission lines in the IRS wavelength range we 
304: defined four major excitation classes. We designate the PNe to be low excitation if 
305: only the [\ion{S}{3}] lines are visible, intermediate excitation when the spectra shows 
306: [\ion{Ar}{3}], [\ion{S}{4}], and [\ion{Ne}{3}] nebular emission, high excitation if we also 
307: detect [\ion{O}{4}], and very high excitation if we also see the [\ion{Ne}{4}] or 
308: [\ion{Ne}{5}] emission. Several PNe show both low and high excitation lines, likely 
309: due to different excitation zones in the nebulae. 
310: 
311: In order to study the emission from the dust, a single blackbody was fitted to the 
312: merged cleaned continuum using data from both nod positions. Nebular lines and 
313: solid-state features were masked during the fitting procedure to isolate the continuum. 
314: Data with $\lambda >38 \mu$m in the LL1 module were often masked as well due to 
315: the increased noise at these wavelengths. Figures 2 through 5 show the best
316: fit Black-body curves to the data, which are adopted as the continuum 
317: fits, and the continuum-subtracted spectra in four PNe representing a variety of IRS spectral features . 
318: The top panels show, for each nod position, the original spectra and the Black-body fit (indicated by a thicker 
319: line); the bottom panel shows the continuum-subtracted spectra and includes the 
320: emission line identifications. 
321: Solid-state features were measured using an IDL routine developed by us which 
322: allows for varying the start and end-points of the feature to be fitted. Solid-state fits 
323: were performed on the continuum-subtracted spectra.
324: 
325: All the IRS spectra show evidence of a thermal continuum with superimposed 
326: nebular emission lines. In addition, the presence of solid-state emission features
327: are superimposed on some of the spectra. 
328: We sort the PNe into three different IRS spectral types: F, CRD, and ORD types. 
329: We define as F (for featureless) those PNe that show no evidence of solid-state features. 
330: The lack of features is 
331: due either to the actual absence of grain emission, such 
332: as may be expected in evolved PNe, or to the dust features possibly being below the 
333: detection limit. In this class, we include the objects that show possible but not certain 
334: dust emission features. 
335: CRD (for carbon-rich dust) PNe are those whose spectra show carbon-rich dust features. This class includes multiple 
336: types of dust grains. Among CRD we find spectra showing the very broad 11 $\mu$m 
337: emission of silicon carbide, 
338: other spectra showing broad peaks at 6-9, 11-15, and 15-21 $\mu$m generally attributed to very small
339: carbonaceous grains, possibly formed by large PAH clusters.
340: PNe with evidence of the classic PAH emission features at 3.3, 6.2, 7.7, 8.6, and/or 
341: 11.3~$\mu$m are classified as CRD. Some CRD PNe show both broad and 
342: narrow carbon-rich dust features, superimposed, as if the ionized PAH features were 
343: beginning to form out of the clusters of small carbonaceous grains as evolution 
344: proceeds. 
345: ORD (for oxygen-rich dust) PNe are those whose spectra show oxygen-rich dust emission features, such as crystalline 
346: silicates. 
347: 
348: In Figures 6 through 8 we show all the observed IRS spectra. Figures 6a, 6b, and 6c include all F
349: PNe (listed alphabetically, the LMC PNe first). 
350: Figures 7a, 7b, and 7c show the CRD
351: PNe. In Figures 7a through 7c we order the PNe by the strengths of the carbonaceous dust features, as they generally 
352: are thought to evolve 
353: in PNe. The sequence shows the evolution of certain features, but it is not an evolutionary sequence
354: since PNe belong to different mass and/or metallicity sequences.
355: Figure 8 shows the small sample of ORD PNe, where PNe have been ordered 
356: according to the strength of the degree of processing of their silicate features. All spectra show the two nod positions
357: averaged. In Figures 7 and 8 these averages were normalized with respect
358: to the flux at 14 $\mu$m, a feature-free region of all spectra.
359: 
360: In Table 3 we summarize the results from the analysis of the IRS spectra. In column (2) 
361: we give the dust temperature of the continuum, fitted as a black body; column (3) 
362: gives the infrared flux derived from the integration of the continuum dust spectra 
363: between 5 and 40~$\mu$m; column (4) gives the IR excess; 
364: column (5) gives their dust type; and 
365: column (6) gives the excitation class of the PN, based on the observed emission 
366: lines. 
367: From Montecarlo simulation based on several fits for the continua we infer that the fitting errors
368: are below $\sim$5$\%$ both in temperatures and IR fluxes.
369: To obtain the ratios in column (4) we have used the Balmer line fluxes of these 
370: PNe from Stanghellini et al. (2002, 2003), and Shaw et al. (2006). 
371: 
372: 
373: Below we list the general characteristics of the observed PNe. The dust type,
374: based 
375: on the infrared spectrum for each target, is included. 
376: 
377: 
378: \noindent{\bf SMP~LMC~4} (Fig.~6a, F): With a low continuum level, the spectrum of this PN is regular
379: and does not show solid-state features with the exception of very faint emission at 
380: 11.3 $\mu$m, which could be of interstellar nature, or the remnant of evaporated dust. The spectrum is almost featureless, 
381: but its classification based on solid-state features is uncertain.
382: The nebular emission lines are characteristic of a very high excitation spectrum.
383: 
384: \noindent{\bf SMP~LMC~9} (Fig.~7b, CRD): This spectrum shows faint emission 
385: at 11.3 $\mu$m, which is likely to be from PAHs. A host of nebular emission lines show a very high excitation
386: nebula by the presence of [Ne V] at 24.2 $\mu$m.
387: 
388: \noindent{\bf SMP~LMC~10} (Fig.~6a, F): A featureless spectrum apart from the nebular emission lines, including
389: high excitation ones.
390: 
391: \noindent{\bf SMP~LMC~16} (Fig.~6a, F): The line at about 9.8 $\mu$m could be molecular 
392: hydrogen at
393: $\lambda$9.66 $\mu$m, which is in contrast with the very high excitation nebular lines. 
394: 
395: \noindent{\bf SMP~LMC~18} (Fig.~6a, F): Faint and featureless, with the exception of a possible
396: broad dust feature at 30$\mu$m.
397: The spectrum shows intermediate excitation nebular lines. Its classification based on dust features is uncertain due to 
398: a possible very low detection of the 15--20$\mu$m broad feature.
399: 
400: \noindent{\bf SMP~LMC~19} (Fig.~7b, CRD): The feature at 11.3 $\mu$m is likely to be narrow PAH emission.
401:  The emission spectrum indicates very high excitation.
402: 
403: \noindent{\bf SMP~LMC~20} (Fig.~6a, F): As in the spectrum of SMP~LMC~16, we are possibly detecting
404: molecular hydrogen at 9.66 $\mu$m. Again, as in SMP~LMC~16, the nebular lines denote a very high excitation PN, thus the
405: molecular line is rather improbable.
406: 
407: \noindent{\bf SMP~LMC~21} (Fig.~8, ORD): This PN has a unique spectrum that shows strong crystalline
408: silicate emission. The nebular emission line spectrum is also 
409: interesting, showing a combination of low (i.e., [S III])
410: excitation nebular emission lines with very high excitation
411:                  ones (i.e., [Ne VI]).
412: 
413: \noindent{\bf SMP~LMC~25} (Fig.~7a, CRD): This spectrum shows carbon-rich dust features, which are likely 
414: due to PAH clusters, or very small carbonaceous grains. We observe peaks at 6.2, 7.7, and 8.5 $\mu$m, all
415: likely to be proto-PAH features, and a huge SiC emission at 11.3 $\mu$m. 
416:  This is an intermediate excitation nebula.
417: 
418: \noindent{\bf SMP~LMC~27} (Fig.~6a, F): This is an intermediate excitation nebula with no dust features, with the possible exception
419: of the 30$\mu$m broad feature as in SMP~LMC~18.
420: 
421: \noindent{\bf SMP~LMC~34} (Fig.~6b, F): An almost featureless spectrum, except for the nebular emission lines
422: typical of intermediate excitation, and the solid-state feature at 30$\mu$m, as in
423: SMP~LMC~18. It is hard to classify it based on solid-state features.
424: 
425: \noindent{\bf SMP~LMC~45} (Fig.~6b, F): The nebular spectrum shows low to high excitation features, 
426: while the solid-state features are missing with the possible exception of the 30$\mu$m feature.
427: 
428: \noindent{\bf SMP~LMC~46} (Fig.~7b, CRD): The features at 7.9 $\mu$m and 11.3 $\mu$m could be PAHs or PAH-related, but very faint.
429: The nebula has intermediate to high excitation.
430: 
431: \noindent{\bf SMP~LMC~48} (Fig.~7a, CRD): The spectrum shows several complex solid state features. The broad feature
432: at 6--9 $\mu$m, possibly due to small grain clusters, is superimposed on narrow features at 6.2, 7.7, and 8.6
433: $\mu$m, characteristic of PAHs. Other features are also seen at 15--20 $\mu$m and 30$\mu$m. 
434: The PN is of intermediate excitation from the nebular
435: lines.
436:  
437: \noindent{\bf SMP~LMC~66} (Fig.~6b, F): In this nebula it is most apparent that the feature at
438: 29.6 $\mu$m is an artifact, since it only appears in one of the nodding
439: positions. This is a high excitation nebula with no solid state features
440: in the spectrum.
441: 
442: \noindent{\bf SMP~LMC~71} (Fig.~7c, CRD): This is a very high excitation spectrum with a
443: solid state feature
444: at 6.2 to 8.6 $\mu$m, possibly due to small grain clusters, superimposed on narrow features at 6.2, 7.7, and 8.6
445: $\mu$m, characteristic of PAHs. Other features are also seen at 15--20 $\mu$m and 30$\mu$m. 
446: 
447: \noindent{\bf SMP~LMC~72} (Fig.~6b, F): This is a very high excitation spectrum without solid-state features.
448: 
449: \noindent{\bf SMP~LMC~79} (Fig.~7c, CRD): Very interesting PAH features are present at 6.2, 7.7, 8.6, 11.3, and 13.5 $\mu$m, 
450: superimposed on broad features. Spectral lines are characteristic of a high
451: excitation PN.
452: 
453: \noindent{\bf SMP~LMC~80} (Fig.~6b, F): Dust featureless spectrum with intermediate excitation emission lines.
454: The thermal continuum is undetected in this PN.
455: 
456: \noindent{\bf SMP~LMC~81} (Fig.~8, ORD): The spectrum presents evident silicate features (SiO) at 9--12.5 $\mu$m.
457: The broad emission between 16 and 30 $\mu$m is probably due to dust continuum. 
458: The nebular emission spectrum shows intermediate excitation lines.
459: 
460: \noindent{\bf SMP~LMC~95} (Fig.~6b, F): This nebula shows a high excitation spectrum. Although almost featureless, the 
461: solid-state spectrum is
462: characteristic of an evolved PN, with a possible faint PAH feature showing at 11.3 $\mu$m.
463: 
464: \noindent{\bf SMP~LMC~97} (Fig.~6c, F): This is a very high excitation PN without solid state features, with the possible exception
465: of the 30$\mu$m broad feature.
466: 
467: \noindent{\bf SMP~LMC~99} (Fig.~7c, CRD): This object shows a very high excitation spectrum. The solid-state features are similar to those
468: of SMP~LMC~71, with narrow PAH emission features superimposed on the broad dust features, possibly due to clusters of
469: small grains. The 30$\mu$m broad feature is evident in the spectrum.
470: 
471: 
472: \noindent{\bf SMP~LMC~100} (Fig.~7c, CRD): A very high excitation PN, this spectrum shows broad PAH signature and the 11--15 $\mu$m bump
473: typical of unevolved carbon-rich dust.
474: 
475: \noindent{\bf SMP~LMC~102} (Fig.~6c, F): There is a possible PAH detection in this otherwise featureless
476: spectrum.
477: 
478: \noindent{\bf SMP~SMC~2} (Fig.~7b, CRD): The spectrum of this very high excitation PN
479: shows very likely PAH emission at 11.3 $\mu$m. 
480: 
481: \noindent{\bf SMP~SMC~5} (Fig.~7c, CRD): Several PAH features are seen in this PN, superimposed on broad carbon-rich dust
482: emission peaking at 8, 11.3 $\mu$m. The nebular excitation is very high. 
483: 
484: \noindent{\bf SMP~SMC~8} (Fig.~6c, F): The spectrum is featureless from the viewpoint of solid-state 
485: compounds. The nebular excitation is intermediate.
486: 
487: \noindent{\bf SMP~SMC~9} (Fig.~6c, F): The spectrum shows no features and no emission lines, and the two nod
488: position do not have the same continuum level. 
489: 
490: \noindent{\bf SMP~SMC~13} (Fig.~7a, CRD): This intermediate excitation spectrum shows broad carbon-rich features with peaks
491: around 8 and 12 $\mu$m, typical of carbon-rich unevolved dust grains, and the broad feature at 30 $\mu$m.
492: 
493: \noindent{\bf SMP~SMC~14} (Fig.~7c, CRD): A high excitation PN, shows narrow PAH features 
494: with peaks at 6.2, 7.7, 8.6, and 11.3, characteristic of PAH emission, superimposed on broad features.
495: 
496: \noindent{\bf SMP~SMC~15} (Fig.~7a, CRD): The [\ion{S}{4}] $\lambda$10.5 emission appears on top of 
497: the broad solid state feature. 
498: This nebula is an extremely good example of the emission expected from carbon-rich dust grains before 
499: they
500: become PAHs. Note that the 15--21 $\mu$m feature may be the precursor of the PAH feature that would
501: be found at 17.0--17.7 $\mu$m. In addition, the broad 30 $\mu$m feature is also clearly seen. 
502: This PN is an intermediate excitation one.
503: 
504: \noindent{\bf SMP~SMC~16} (Fig.~7b, CRD): 
505: Broad carbon-rich dust features appear with peaks at 6--9 and 11-15  $\mu$m, typical of unprocessed clusters 
506: of small carbonaceous grains and/or large PAH clusters. This is a low excitation PN. 
507:  
508: \noindent{\bf SMP~SMC~17} (Fig.~7c, CRD): Similar to SMP~SMC~16, but with PAH narrow peaks appearing 
509: at 6.2, 7.7, 8.7, and 11.3 $\mu$m.
510: This is an intermediate excitation PN. 
511: 
512: \noindent{\bf SMP~SMC~18} (Fig.~7a, CRD): This is another very good example of the emission 
513: expected in a carbon-rich source with dust grains where PAHs have not yet been formed, showing broad SiC emission. 
514: Note the plateau features at 6--9 $\mu$m, 11-15 $\mu$m, and 15--21 $\mu$m, as well as the 
515: extremely broad 30 $\mu$m feature. The excitation for this nebula is intermediate.
516: 
517: \noindent{\bf SMP~SMC~19} (Fig.~7c, CRD): Some carbon-rich dust features, both narrow PAH peaks and broad, unevolved ones, 
518: are present, together with very high excitation lines.
519: 
520: \noindent{\bf SMP~SMC~20} (Fig.~7a, CRD): This source is similar to SMP~SMC~15; with huge SiC features, it is probably
521: the least evolved source among the PNe showing carbon-rich features.
522: 
523: \noindent{\bf SMP~SMC~23} (Fig.~6c, F): This object has a featureless spectrum, apart from a couple of 
524: intermediate excitation emission lines. 
525: 
526: \noindent{\bf SMP~SMC~25} (Fig.~8, ORD): Unique source in the SMC showing oxygen-rich dust silicates peaking 
527: shortward of 11 $\mu$m. The solid state features are similar to those of 
528: SMP~LMC~81, but with some crystalline silicates features arising over 
529: the continuum longward of 20 $\mu$m. This is a very high excitation 
530: nebular spectrum, high
531: dust and effective temperature, and a very high central star mass PN.
532: 
533: \noindent{\bf SMP~SMC~26} (Fig.~6c, F): The faint feature at 11.3 $\mu$m could be a PAH. The nebular spectrum shows very
534: high excitation lines.
535:      
536: \noindent{\bf SMP~SMC~27} (Fig.~7b, CRD): Intermediate excitation PN with carbon-rich dust emission of the broad type.
537: 
538: 
539: 
540: \section{Correlations between IR spectral characteristics and other nebular and stellar properties}
541: 
542: We studied the interrelations between the properties that we derived from the IRS spectra and the 
543: properties derived from other studies. We have specifically selected a target sample of PNe whose 
544: morphology, gaseous carbon abundance, and central stars have been studied
545: through {\it HST} imaging and spectroscopy. The sizes of the PNe have been determined from the 
546: angular size measured from {\it HST} images (see Table 1) and the distances to the Magellanic Clouds. In this paper we 
547: adopt d$_{\rm LMC}$ =50.6 kpc (Freedman et al. 2001; Mould et al. 2000)
548: and d$_{\rm SMC}$ = 58.3 kpc (Westerlund 1997), used also to get the IR 
549: luminosity in solar units from the fluxes of Table~3.
550: 
551: The PNe in our sample represent an assortment of initial masses and metallicities; they show a variety
552: of morphological types and also represent a variety of 
553: evolutionary stages. Care is needed in disentangling the evolutionary effects on the physical quantities 
554: plotted and their intrinsic values, related to mass and metallicity. 
555: 
556: 
557: Figure 9 shows the measured infrared luminosity versus carbon abundance (here the abundances are in terms of log X/H +12,
558: where X is the element considered)
559: of all PNe of our sample that have a carbon determination either from Leisy \& Dennefeld (1996)
560: or from Stanghellini et al. (2005). We estimated that the infrared luminosity errors are of the order of 5$\%$;
561: typical carbon abundance uncertainties are $\sim5$ to 10$\%$. This implies that errorbars in Figure 9
562: would be smaller or comparable to the sizes of the symbols used for the data points. 
563: The two parts of Figure 9 display PNe on the same plane, 
564: where dust type and morphology are identified by symbol type in the left and right panels respectively.
565: The loci of the different dust type PNe are well separated in this
566: plot: the PNe with featureless spectra (F), are typically at low to intermediate luminosity and present 
567: a spread in carbon
568: abundances. The PNe with CRD spectra are on the upper right of the
569: plot. PNe with oxygen-rich dust spectra, or ORD, are on the upper left of the plot. From the figure 
570: we infer a
571: strong relation between the gas and dust composition: CRD IRS spectra are only
572: found in carbon-rich PNe. 
573: 
574: 
575: By comparing the left and right panels of Figure 9
576: we note that no bipolar, quadrupolar, or point-symmetric PNe (squares)
577: in the figure correspond to PNe with carbonaceous dust. The carbon
578:  gas abundance, dust type, and morphology are thus very clearly interrelated.
579: Since gaseous carbon is depleted in massive PN progenitors via the HBB reactions, 
580: in Figure 9 the progenitor mass of PNe decreases from left to right. 
581:  PNe with high mass progenitors
582:  in the figure are ORD PNe if they are IR-bright;
583:  otherwise they are featureless. It is possible that some of the F PNe on the lower left of the plot
584:  are indeed evolved versions of the ORD PNe on the upper left. 
585:  It is worth noting that the SMC PNe in the plot are restricted to the upper right part, with log C/H +12$>$8.4 and
586:  log L$_{\rm IR}$/L$_{\odot}>$1.9.
587:  
588:  More insight comes from Figure 10, where we plot the carbon vs. oxygen
589: abundances (oxygen abundance uncertainties are similar or smaller than
590: those for carbon, thus
591: errorbars in Figure 10 are smaller than symbol size). Once again, the left and right panels identify dust type and morphology.
592: The left panel shows that PNe with PAHs and other 
593:  carbonaceous dust features all have C/O$>$1, while C/O$<$1 for all
594:  ORD PN, and there are no exceptions to these correlations. 
595: The right panel of Figure 10
596:  shows that all asymmetric PNe have low carbon abundance, and either oxygen-rich dust or
597: no dust features. It is clear that PNe whose progenitors have gone through the HBB
598: do not develop carbonaceous dust. We found that the average N/O for CRD PNe is $\sim$0.3, while
599: it is $\sim$2 for ORD PNe, a clear sign of HBB activity. 
600: 
601: In Table 4 we summarize the group properties of the different dust types, separating the columns for the LMC and the SMC 
602: PNe. In this table, abundances (Stanghellini 2006) are reported linearly, in terms of hydrogen. 
603: In column 1 we give the diagnostics,
604: and columns (2) through (7) give respectively their values (and standard deviation, in row 2 through 6) for the 
605: samples. Note that in a few cases we do not have data to fill the tables, and in those cases where we did not report 
606: the deviation 
607:  we only have one datum for the diagnostics.
608: It is worth mentioning that ORD PNe, those whose progenitors are likely more massive and which have gone through 
609: HBB nucleosynthesis, are
610: not necessarily those with the higher oxygen abundances. This is a phenomenon that is especially marked in 
611: the Magellanic Cloud PNe, where the enhanced efficiency of the 
612: ON cycle at low metallicity depletes both carbon and oxygen in favor of nitrogen (Stanghellini 2006). This is also why 
613: the
614: oxygen abundances are a poor indicator of PN metallicity for PNe whose progenitors have gone through the 
615: HBB. 
616: 
617: 
618: In Figure 11 we show the dust temperature vs. physical radius of the 
619: nebulae (both quantities have uncertainties smaller than 0.02 dex).
620: The straight line shows the best fit using all plotted data, and suggests a 
621: power-law decline of T$_{\rm dust} \propto$~R$^{-0.25}$. 
622: Once again, dust types are identified in the left panel, morphology in the right panel. 
623: The physical radii of these PNe are photometric radii (e.g., Shaw et al.\ 2001), 
624: and correspond to the geometrical radius of the ionized gas for all except the bipolar
625: PNe with large lobes. Planetary nebulae with larger radii are generally more evolved 
626: than those with small radii, thus it is not surprising that the radii are inversely 
627: proportional to some power of the dust temperature. Lenzuni et al. (1989) showed 
628: a similar plot for Galactic PNe that were observed with the IRAS satellite and 
629: inferred that T$_{\rm dust} \propto~$R$^{-0.15}$. 
630: We confirm that the grain efficiency of PNe (within the same dust type) decreases with 
631: time, probably due to sputtering of material off the grain surface. It is also
632: interesting to note that our much higher precision in the radius determination (because in 
633: Magellanic Cloud PNe, the distances are known and accurate angular
634: sizes are available from the HST imagery) and the dust temperature (since temperature is derived from 
635: a spectral fit, and not by interpolation of a few photometric data points, as for the IRAS 
636: data for Galactic PNe) does not significantly decrease the scatter of the relation compared 
637: to the Galactic plot of Lenzuni et al. (1989); this suggests that the scatter is real and depends 
638: on the grain properties. 
639: 
640: 
641: In Figure 12 we show the tight relation between the dust temperature and the IR luminosity. PNe evolve
642: from high to low dust temperatures, and from high to low IR luminosity, i.e.,
643: from the upper right to the lower left of the plots. All ORD PNe are at intermediate 
644: evolutionary stages. Uncertainties of the data points are smaller than 0.02 dex in
645: both axes. 
646: At very low IR luminosity and temperatures, where the most evolved PNe are found,
647: we only find F PNe. 
648: 
649: The electron density of the gas is another parameter that decreases as the nebula evolves. This is studied 
650: in Figure 13, where the dust temperature and the electron density show a tight correlation.
651:  It is worth noting that the CRD PNe follow a tighter correlation than F and ORD PNe.
652: In is also important to mention that dust temperature does not correlate with optical extinction. Typical errors
653: in the logarithmic density scale are smaller than $\sim$0.1 dex, while the dust
654: temperature uncertainties are negligible in this scale.
655: 
656: 
657: While the anti-correlation of dust temperature with nebular radius indicates a general evolutionary trend, it is only by
658: studying nebulae and their central stars together that one can disentangle the evolutionary effects from the
659: effect of the mass of the PN progenitors and the final core mass of the central stars. In the HR diagram of Figure 14
660: we can
661: characterize the evolution of post-AGB stars by mass bins, thus singling out the effects of evolution from those
662: due to dynamical evolution and nebular acceleration.
663: 
664: In Figure 14 we show the loci of the central stars, of those PNe in our sample whose stellar physical
665: parameters are available, on the log T$_{\rm eff}$--logL/L$_{\odot}$ plane
666: (individual effective temperature errors
667: are negligible on the log scale (see Villaver et al. 2003, 2004, 2006), 
668: while luminosity uncertainties are typically
669: of the order of 0.05 dex).
670: The plot shows the evolutionary tracks 
671: of hydrogen-burning post-AGB stars with masses of 0.57, 0.6, 0.7, 0.,75, and 0.9 M$_{\odot}$, taken from
672: the models of Stanghellini \& Renzini (2000). The 
673: symbols refer to the dust type 
674: (left panel) and morphological (right panel). We have plotted only stars whose luminosity has been measured directly, 
675: from their magnitudes, and have not
676: included model-dependent luminosity determination (Villaver et al. 2003, 2004, 2006). We only used
677: those stars whose effective temperatures have been evaluated via He II Zanstra analysis. From the figure we infer that
678: the F and CRD PNe can be found around stars with a
679: range of 
680: evolutionary stages and core masses, while the only ORD PNe whose location on the HR diagram
681: is known correspond to a high mass evolutionary track.
682: 
683: 
684: In Figure 15 we show the sequence of CRD PNe with nebular size. All PNe whose spectra have been depicted in Figure 7
685: (a through b) have been assigned a sequence number, 1 to 20, as ordered in the Figure. The sequence starts with PNe 
686: featuring huge SiC
687: (in order of strength), then goes to those PNe with broad and narrow carbonaceous features superimposed, then 
688: to the PAH feature PNe, and finally to those PNe with strong PAH emission. 
689: The symbols in Figure 15 refer to nebular morphology with the usual coding. We infer that the nebular
690: expansion is related to the carbonaceous dust sequence, linking nebular evolution and dust
691: evolution in particular for round and elliptical PNe. 
692: The scatter represents the PN spread in progenitor mass and metallicity.
693: 
694: 
695: 
696: \section{Discussion}
697: 
698: In our analysis we found that 56\% of the Magellanic Cloud PNe studied present solid state features, 
699: superimposed on the thermal dust continuum. These solid state features are, in 87\% of cases, the 
700: signature of carbonaceous dust, with only a few PNe showing oxygen-rich dust emission instead.
701: The rest of the nebulae show only nebular emission lines on a thermal continuum, in some cases 
702: very low.
703: By contrast, such a low continuum and absence of features has
704: not been observed in Galactic PNe (Garc{\'{\i}}a-Lario et al., in preparation).
705: It is remarkable that in the low-metallicity environments of the Magellanic Clouds
706: the solid state features seem to be rarer than in Galactic PNe.
707: This simple observation shows that the metallicity of the population has an enormous impact 
708: on the dust that forms on the AGB, as we knew already from carbon star counts in the Local Group Galaxies. What is
709: new here is that the nature of the dust in PNe is different for different environments. Mass-loss, if caused by pressure 
710: on the dust grains, must vary accordingly. The selection effects are different for Galactic and Magellanic Cloud PNe,
711: and a quantitative comparison of the Magellanic Cloud and Galactic samples is beyond the scope of this paper, but it
712: is important to note that the selection effects that affect the LMC and SMC samples are comparable,
713: thus the two samples readily comparable.
714: 
715: The ratio of CRD versus ORD PNe is 11 in the SMC, and 4.5 in the LMC, compared to $\sim$1 in the Galaxy
716: (Garc{\'{\i}}a-Lario et al., in preparation).
717: Selection effects limit the quantitative impact of the comparison between 
718: Magellanic Cloud PNe and the Galactic results.
719: The evolution of carbonaceous dust proceeds from the amorphous state, to crystalline, to PAH; the grains
720: evolve from small clusters of molecules to ionized PAH, as the central star becomes hotter (Peeters et al. 2004;
721: Moutou et al. 2000). The grains become bigger
722: but eventually evaporate under the action of the UV radiation field; thus, at late evolutionary stages
723: we would not observe much dust in PNe. 
724: Not only is the dust chemistry different in different environments, but the evolutionary stage of the 
725: dust grains is also statistically different. For example, the SiC compounds are found only in SMC PNe and not
726: in the LMC (or their Galactic) counterparts. In our sample
727: there are three SMC PNe (SMP~SMC~15, SMP~SMC~18, and SMP~SMC~20)
728: with broad SiC emission, but none in the LMC. Given that the LMC sample is larger than the SMC sample,
729: this is remarkable and of course needs to be studied with larger samples in both galaxies.
730: A few other SMC and LMC PNe have broad carbonaceous dust emission, such
731: as SMP~SMC~13, SMP~LMC~25, and SMP~LMC~48. 
732: Overall, features of amorphous dust grains are more
733: common in the SMC than in the LMC PNe and are very uncommon in the Galaxy. 
734: If these structures really are the early evolutionary stages
735: of PAHs, then we have found that the PN evolutionary rate depends strongly on population metallicity.
736: 
737: 
738: We found that there are strong correlations between the dust type, the carbon abundance, and the
739: IR integrated luminosity. CRD and ORD PNe are generally IR-bright, and at the 
740: opposite extremes of the carbon abundance range. All CRD PNe have C/O$>$1, and 
741: all ORD PNe have C/O$<$1. 
742: Most of the F PNe are low IR-luminous objects, suggesting that 
743: some of them might be an evolved version of the other type of PNe. 
744: All CRD PNe are either round or elliptical in shape, some with a bipolar core, but none of them show
745:  extended lobes or other marked asymmetries. The opposite is true for ORD PNe.
746:  
747: The dust temperature correlates well with other physical parameters that change with evolution,
748:  (e.g., the nebular radius, the IR-luminosity, and the electron density). CRD PNe define a tight sequence
749:  on the log T$_{d}$--log R plane (where R is the photometric radii of the PNe) similar to those found in Galactic PNe. 
750:  The CRD PNe whose spectra
751:  shows amorphous carbonaceous dust grains, such as SiC, have small physical radii,
752:  high electron density, and high IR-luminosity, in agreement with their being the unevolved version of PNe 
753:  with PAH emission. Our empirical evolutionary sequence of carbonaceous dust features correlates well with the
754:  physical dimension of the nebulae, and their dust temperatures. Other correlations of IRS spectral type with
755:  excitation classes, and a modeling approach to these data will be thoroughly discussed in a future
756:  paper (Stanghellini et al., in preparation).
757:  
758:  We have placed the central stars of the nebulae studied on the HR diagram and compared the central star locus with the
759:  evolutionary tracks for these post-AGB stars. We find strong systematics relating the dust type and
760:  the loci of their central stars on the HR diagram. For example, most F PNe are hosted by evolved central stars, 
761:  their
762:  dust features having probably disappeared due to the UV radiation field of their central stars. We also find that 
763:  central stars of CRD PNe are spread through the diagram, with the uncertain classified at low luminosity, and the 
764:  only SiC PNe for which we have the HR locus is in the early evolutionary stage for the central star as well.
765:   The last row of Table 4 clearly hints that
766:  CRD PNe have central stars that are typical for the galaxy they belong to 
767:  (see Villaver et al. 2006), while
768:  the only ORD PNe with measured mass is very massive. We need several
769:  more data points to substantiate this finding.
770:  
771:  Our observations show that there are fundamental
772:   differences in the characteristics of the grains formed around
773:   PNe at different metallicities, which are a consequence of the 
774:   more inefficient production of dust, the smaller mass-loss rates
775:   and slower evolution of the central star experienced by PNe at
776:   low metallicities compared to their Galactic analogs. At the low metallicities of the Magellanic Clouds
777: fewer grains are formed, with respect to their Galactic counterparts. The dust features that we see in our targets are
778: of moderate strengths.
779: 
780: \section{Conclusions}
781: 
782: The IRS spectra of a selection of Large and Small Magellanic Cloud PNe were analyzed for their dust content to
783: determine the nature of their dust chemistry and to gain knowledge of their dust evolutionary phase.
784: The spectra studied show three major dust emission profiles, and we classify then into:
785: F, those dominated by nebular emission lines; CRD, whose spectra show
786: carbonaceous dust; and ORD, with oxygen-rich dust compounds seen in emission.
787: 
788: The dust properties, including dust type, IR luminosity, and dust temperatures, do correlate strongly with the
789: known gas properties of the PNe. We find an exclusive correlation between dust type and carbon abundance, showing that
790: {\it all CRD PNe have C/O$>$1, and all ORD PNe have gone through the HBB phase}. We also find an exclusive 
791: correlation between CRD PNe and symmetric morphology, while all ORD PNe show high asymmetries. 
792: We also find hints of more massive central stars in ORD PNe.
793: 
794: The dust temperature decreases with increasing physical radii, and with decreasing electron density.
795: CRD PNe define a scattered evolutionary sequence, the scatter being due to the range of initial masses and
796: metallicity within the sample.
797: 
798: The statistics of dust emission features that we see in the Magellanic Clouds, if compared with 
799: that of the Galaxy, indicate that low metallicity environments do not favor the production of dust. As a consequence, 
800: the 
801: mass loss must be lower if the mass ejection is due to pressure on the dust grains. In particular, 
802: the spectacular bipolar 
803: nebulae, if indeed more massive and evolved from massive progenitors, are rarer in the Magellanic Clouds than in 
804: the galaxy, and the findings of the dust properties of this work conform to this scenario.
805: 
806: The PN sample analyzed in this paper represents an important fraction of all the LMC and SMC PNe ever
807: observed with {\it HST}. Further {\it HST} and Spitzer observations on a larger Magellanic 
808: Cloud PN sample, such as shown in the outer contours of Figure 1,  would be a good complement 
809: to this study. While the present Spitzer capabilities limit spectroscopic studies 
810: to bright Magellanic Cloud PNe, future space technology might 
811: extend the analysis to VLE PNe, such as those found in recent surveys 
812: (Reid \& Parker 2006ab). 
813: 
814: A comparative study of the LMC and SMC PNe together with those in the Galaxy, based on their mid-IR spectra, 
815: is also planned for the future. 
816: 
817: 
818: 
819: \acknowledgments
820: 
821: Support for this work was provided by NASA through a grant issued by JPL/Caltech.
822: Arturo Manchado, Letizia Stanghellini, and Eva Villaver are grateful to the European Space 
823: Astronomy Centre for their hospitality during two science group meetings. 
824: Arturo Manchado acknowledges support from grant \emph{AYA 2004$-$3136}
825: from the Spanish Ministerio de Educaci\'on y Ciencia. We warmly thank Terry Mahoney for 
826: carefully reading the manuscript. We acknowledge the input from an anonymous referee.
827: 
828: \begin{thebibliography}{}
829: \bibitem[Caputo et al.(1999)]{1999ApJ...525..784C} Caputo, F., Marconi, M., 
830: \& Ripepi, V.\ 1999, \apj, 525, 784 
831: \bibitem[Cioni et al.(2003)]{2003A&A...406...51C} Cioni, M.-R.~L., et al.\ 
832: 2003, \aap, 406, 51 
833: \bibitem[Cioni \& Habing(2003)]{2003A&A...402..133C} Cioni, M.-R.~L., \& 
834: Habing, H.~J.\ 2003, \aap, 402, 133 
835: \bibitem[Cohen \& Barlow(1974)]{1974ApJ...193..401C} Cohen, M., \& Barlow, 
836: M.~J.\ 1974, \apj, 193, 401 
837: \bibitem[Freedman et al.(2001)]{2001ApJ...553...47F} Freedman, W.~L., et 
838: al.\ 2001, \apj, 553, 47 
839: \bibitem[Groenewegen et al.(2000)]{2000MmSAI..71..639G} Groenewegen, 
840: M.~A.~T., Blommaert, J.~A.~D.~L., Cioni, M.-R., Okumura, K., Habing, H.~J., 
841: Trams, N.~R., \& van Loon, J.~T.\ 2000, Memorie della Societa Astronomica 
842: Italiana, 71, 639 
843: \bibitem[Garc{\'{\i}}a-Hern{\'a}ndez et al.(2007)]{2007A&A...462..711G} 
844: Garc{\'{\i}}a-Hern{\'a}ndez, D.~A., Garc{\'{\i}}a-Lario, P., Plez, B., 
845: Manchado, A., D'Antona, F., Lub, J., \& Habing, H.\ 2007, \aap, 462, 711 
846: \bibitem[Garc{\'{\i}}a-Lario \& Perea 
847: Calder{\'o}n(2003)]{2003eida.conf...97G} Garc{\'{\i}}a-Lario, P., \& Perea 
848: Calder{\'o}n, J.~V.\ 2003, ESA SP-511: Exploiting the ISO Data 
849: Archive.~Infrared Astronomy in the Internet Age, 97 
850: \bibitem[]{986} Houck, J. R., et al. 2004, ApJS, 154, 18
851: \bibitem[Jacoby \& De Marco(2002)]{2002AJ....123..269J} Jacoby, G.~H., \& 
852: De Marco, O.\ 2002, \aj, 123, 269 
853: \bibitem[Kaeufl et al.(1993)]{1993ApJ...410..251K} Kaeufl, H.~U., Renzini, 
854: A., \& Stanghellini, L.\ 1993, \apj, 410, 251 
855: \bibitem[Leisy \& Dennefeld(1996)]{1996A&AS..116...95L} Leisy, P., \& 
856: Dennefeld, M.\ 1996, \aaps, 116, 95 
857: \bibitem[Lenzuni et al.(1989)]{1989ApJ...345..306L} Lenzuni, P., Natta, A., 
858: \& Panagia, N.\ 1989, \apj, 345, 306 
859: \bibitem[Marigo(2001)]{2001A&A...370..194M} Marigo, P.\ 2001, \aap, 370, 
860: 194 
861: \bibitem[Mould et al.(2000)]{2000ApJ...529..786M} Mould, J.~R., et al.\ 
862: 2000, \apj, 529, 786 
863: \bibitem[Moutou et al.(2000)]{2000A&A...354L..17M} Moutou, C., Verstraete, 
864: L., L{\'e}ger, A., Sellgren, K., \& Schmidt, W.\ 2000, \aap, 354, L17 
865: \bibitem[Peeters et al.(2004)]{2004ApJ...617L..65P} Peeters, E., Mattioda, 
866: A.~L., Hudgins, D.~M., \& Allamandola, L.~J.\ 2004, \apjl, 617, L65 
867: \bibitem[Peimbert \& Serrano(1980)]{1980RMxAA...5....9P} Peimbert, M., \& 
868: Serrano, A.\ 1980, Revista Mexicana de Astronomia y Astrofisica, 5, 9 
869: \bibitem[Reid \& Parker(2006a)]{2006MNRAS.373..521R} Reid, W.~A., \& Parker, 
870: Q.~A.\ 2006, \mnras, 373, 521 
871: \bibitem[Reid \& Parker(2006b)]{2006MNRAS.365..401R} Reid, W.~A., \& Parker, 
872: Q.~A.\ 2006, \mnras, 365, 401 
873: \bibitem[Shaw et al.(2001)]{2001ApJ...548..727S} Shaw, R.~A., Stanghellini, 
874: L., Mutchler, M., Balick, B., \& Blades, J.~C.\ 2001, \apj, 548, 727 
875: \bibitem[Shaw et al.(2006)]{2006ApJS..167..201S} Shaw, R.~A., Stanghellini, 
876: L., Villaver, E., \& Mutchler, M.\ 2006, \apjs, 167, 201 
877: \bibitem[Schultheis et al.(2004)]{2004A&A...427..945S} Schultheis, M., 
878: Glass, I.~S., \& Cioni, M.-R.\ 2004, \aap, 427, 945 
879: \bibitem[Stanghellini et al.(1999)]{1999ApJ...510..687S} Stanghellini, L., 
880: Blades, J.~C., Osmer, S.~J., Barlow, M.~J., \& Liu, X.-W.\ 1999, \apj, 510, 
881: 687 
882: \bibitem[Stanghellini \& Renzini(2000)]{2000ApJ...542..308S} Stanghellini, 
883: L., \& Renzini, A.\ 2000, \apj, 542, 308 
884: \bibitem[Stanghellini et al.(2002)]{2002ApJ...575..178S} Stanghellini, L., 
885: Shaw, R.~A., Mutchler, M., Palen, S., Balick, B., \& Blades, J.~C.\ 2002, 
886: \apj, 575, 178 
887: \bibitem[Stanghellini et al.(2003)]{2003ApJ...596..997S} Stanghellini, L., 
888: Shaw, R.~A., Balick, B., Mutchler, M., Blades, J.~C., \& Villaver, E.\ 
889: 2003, \apj, 596, 997 
890: \bibitem[Stanghellini et al.(2005)]{2005ApJ...622..294S} Stanghellini, L., 
891: Shaw, R.~A., \& Gilmore, D.\ 2005, \apj, 622, 294 
892: \bibitem[Stanghellini(2006)]{2006astro.ph.10206S} Stanghellini, L.\ 2006, 
893: ArXiv Astrophysics e-prints, arXiv:astro-ph/0610206 
894: \bibitem[Trams et al.(1999)]{1999A&A...346..843T} Trams, N.~R., et al.\ 
895: 1999, \aap, 346, 843 
896: \bibitem[Vassiliadis \& Wood(1993)]{1993ApJ...413..641V} Vassiliadis, E., 
897: \& Wood, P.~R.\ 1993, \apj, 413, 641 
898: \bibitem[Villaver et al.(2002)]{2002ApJ...581.1204V} Villaver, E., 
899: Manchado, A., \& Garc{\'{\i}}a-Segura, G.\ 2002, \apj, 581, 1204 
900: \bibitem[Villaver et al.(2006)]{2006astro.ph.10079V} Villaver, E., 
901: Stanghellini, L., \& Shaw, R.~A.\ 2007, \apj, 656, 831
902: \bibitem[Villaver et al.(2004)]{2004ApJ...614..716V} Villaver, E., 
903: Stanghellini, L., \& Shaw, R.~A.\ 2004, \apj, 614, 716 
904: \bibitem[Villaver et al.(2003)]{2003ApJ...597..298V} Villaver, E., 
905: Stanghellini, L., \& Shaw, R.~A.\ 2003, \apj, 597, 298 
906: \bibitem[Volk(2003)]{2003IAUS..209..281V} Volk, K.\ 2003, IAU Symposium, 
907: 209, 281 
908: \bibitem[]{1039} Werner, M., et al. 2004, ApJS, 154, 1
909: \bibitem[Westerlund(1997)]{1997macl.book.....W} Westerlund, B.~E.\ 1997, 
910: Book
911: \bibitem[Willson(2000)]{2000ARA&A..38..573W} Willson, L.~A.\ 2000, \araa, 
912: 38, 573 
913: \bibitem[Zhang \& Kwok(1991)]{1991A&A...250..179Z} Zhang, C.~Y., \& Kwok, 
914: S.\ 1991, \aap, 250, 179 
915: 
916: 
917: \end{thebibliography}
918: 
919: 
920: \begin{deluxetable}{lrll}
921: 
922: \tablecaption{Target Selection}
923: 
924: \tablehead{
925: \colhead{Object} & 
926: \colhead{R [$\arcsec]$}& 
927: \colhead{Morphology}&
928: \colhead{notes} \\
929: 
930: 
931: }
932: 
933: \startdata
934:   
935: SMP~LMC~4 & 0.69& Elliptical& Faint halo\\
936: SMP~LMC~9 & 0.46& Bipolar core& Barrel shape\\
937: SMP~LMC~10 & 0.88& Pointsymmetric& Spiral shape\\
938: SMP~LMC~16 & 1.00& Bipolar& \\
939: SMP~LMC~18 & 0.51& Round& Possibly Bipolare core\\
940: SMP~LMC~19 & 0.41& Bipolar core& Ring\\
941: SMP~LMC~20 & 0.41& Bipolar core& \\ %S99
942: SMP~LMC~21 & 0.58& Quadrupolar&\\%S99
943: SMP~LMC~25 & 0.23& Round&\\
944: SMP~LMC~27 & 0.46& Quadrupolar& Outer arc and halo\\
945: SMP~LMC~34 & 0.32& Elliptical& \\
946: SMP~LMC~45 & 0.72& Bipolar& Questionable morphology\\
947: SMP~LMC~46 & 0.31& Bipolar core& Possible ring\\
948: SMP~LMC~48 & 0.20& Elliptical& \\
949: SMP~LMC~66 & 0.54& Elliptical& \\%S99
950: SMP~LMC~71 & 0.31& Elliptical&\\
951: SMP~LMC~72 & $\dots$    & Bipolar& \\%S99
952: SMP~LMC~79 & 0.22& Bipolar core& \\
953: SMP~LMC~80 & 0.21& Round& \\
954: SMP~LMC~81 & 0.15& Round& Barely resolved\\
955: SMP~LMC~95 & 0.46& Bipolar core& Ansae\\
956: SMP~LMC~97 & 0.59& Round& \\ %S99
957: SMP~LMC~99 & 0.41& Bipolar core&\\ %ref?
958: SMP~LMC~100 & 0.71& Bipolar core& Possibly quadrupolar\\
959: SMP~LMC~102 & 0.71& Round& Possibly Bipolar core\\
960: 
961: 
962: &&&\\
963: SMP~SMC~2 & 0.25& Round \\
964: SMP~SMC~5 & 0.31& Round \\ %S99
965: SMP~SMC~8 & 0.23&  Round&  \\
966: SMP~SMC~9 &  0.55& Round&  Inner structure \\
967: SMP~SMC~13 & 0.19&  Round& \\
968: SMP~SMC~14 & 0.42&  Round&  Ansae and/or inner structure\\
969: SMP~SMC~15 & 0.17& Round&\\
970: SMP~SMC~16 & 0.18& Elliptical& \\
971: SMP~SMC~17 & 0.25&  Elliptical& Faint detached halo \\
972: SMP~SMC~18 & 0.15&  $\dots$& Unresolved \\
973: SMP~SMC~19 & 0.30&  Round& Outer structure \\
974: SMP~SMC~20 & 0.15& $\dots$ & Unresolved \\
975: SMP~SMC~23 & 0.30&  Bipolar core&  \\
976: SMP~SMC~25 & 0.19& Elliptical&  \\
977: SMP~SMC~26 & 0.28& Pointsymmetric&  \\
978: SMP~SMC~27 & 0.23& Round& Attached outer halo\\
979: \\
980: 
981:    
982: \enddata 
983: \end{deluxetable} 
984: 
985: 
986: 
987: \begin{deluxetable}{llllll}
988: 
989: \tablecaption{Observing log}
990: 
991: \tablehead{
992: \colhead{Object} & 
993: \colhead{R.A.}& 
994: \colhead{DEC}& 
995: \colhead{Obs. date} & 
996: \colhead{ID number} & 
997: \colhead{IRS campaign} \\
998: 
999: \colhead{}& 
1000: \colhead{J2000.0}& 
1001: \colhead{J2000.0} & 
1002: \colhead{}& 
1003: \colhead{}&
1004: \colhead{}\\
1005: }
1006: 
1007: \startdata
1008:   
1009: SMP~LMC~4 & 04$^{h}$43$^{m}$21$^{s}$.50 & $-$71$^{\rm o}$30$'$09$''$.5 & 08/14/2005 & 14703360 & 23.2	\\
1010: SMP~LMC~9 & 04$^{h}$50$^{m}$24$^{s}$.71 & $-$68$^{\rm o}$13$'$17$''$.0 & 09/13/2005 & 14705920 & 24	\\
1011: SMP~LMC~10 & 04$^{h}$51$^{m}$08$^{s}$.90 & $-$68$^{\rm o}$49$'$05$''$.8 & 09/13/2005 & 14700032 & 24	\\
1012: SMP~LMC~16 & 05$^{h}$02$^{m}$01$^{s}$.91 & $-$69$^{\rm o}$48$'$54$''$.4 & 08/14/2005 & 14700288 & 23.2	\\
1013: SMP~LMC~18 & 05$^{h}$03$^{m}$42$^{s}$.64 & $-$70$^{\rm o}$06$'$47$''$.8 & 08/14/2005 & 14700544 & 23.2	\\
1014: SMP~LMC~19 & 05$^{h}$03$^{m}$41$^{s}$.30 & $-$70$^{\rm o}$13$'$53$''$.6 & 08/14/2005 & 14700800 & 23.2	\\
1015: SMP~LMC~20 & 05$^{h}$04$^{m}$40$^{s}$.14 & $-$69$^{\rm o}$21$'$39$''$.3 & 08/14/2005 & 14701056 & 23.2	\\
1016: SMP~LMC~21 & 05$^{h}$04$^{m}$51$^{s}$.90 & $-$68$^{\rm o}$39$'$10$''$.0 & 08/14/2005 & 14701312 & 23.2	\\
1017: SMP~LMC~25 & 05$^{h}$06$^{m}$24$^{s}$.00 & $-$69$^{\rm o}$03$'$19$''$.2 & 08/14/2005 & 14701568 & 23.2	\\
1018: SMP~LMC~27 & 05$^{h}$07$^{m}$54$^{s}$.90 & $-$66$^{\rm o}$57$'$46$''$.1 & 09/13/2005 & 14701824 & 24	\\
1019: % SMP~LMC~30 & 05$^{h}$09$^{m}$19$^{s}$.00 & $-$66$^{\rm o}$53$'$42$''$.0 & 07/14/2005 & 14702080 & 22	\\
1020: SMP~LMC~34 & 05$^{h}$10$^{m}$17$^{s}$.18 & $-$68$^{\rm o}$48$'$23$''$.0 & 08/14/2005 & 14702336 & 23.2	\\
1021: SMP~LMC~45 & 05$^{h}$19$^{m}$21$^{s}$.00 & $-$66$^{\rm o}$58$'$13$''$.0 & 07/14/2005 & 14702592 & 22	\\
1022: SMP~LMC~46 & 05$^{h}$19$^{m}$29$^{s}$.72 & $-$68$^{\rm o}$51$'$09$''$.1 & 08/11/2005 & 14702848 & 23.2	\\
1023: SMP~LMC~48 & 05$^{h}$20$^{m}$09$^{s}$.66 & $-$69$^{\rm o}$53$'$39$''$.2 & 08/14/2005 & 14703104 & 23.2	\\
1024: SMP~LMC~66 & 05$^{h}$28$^{m}$41$^{s}$.20 & $-$67$^{\rm o}$33$'$39$''$.0 & 08/11/2005 & 14703616 & 23.2	\\
1025: SMP~LMC~71 & 05$^{h}$30$^{m}$33$^{s}$.22 & $-$70$^{\rm o}$44$'$38$''$.4 & 08/11/2005 & 14703872 & 23.2	\\
1026: SMP~LMC~72 & 05$^{h}$30$^{m}$45$^{s}$.98 & $-$70$^{\rm o}$50$'$16$''$.4 & 08/11/2005 & 14704128 & 23.2	\\
1027: SMP~LMC~79 & 05$^{h}$34$^{m}$08$^{s}$.76 & $-$74$^{\rm o}$20$'$06$''$.6 & 10/10/2005 & 14704384 & 25	\\
1028: SMP~LMC~80 & 05$^{h}$34$^{m}$38$^{s}$.87 & $-$70$^{\rm o}$19$'$56$''$.9 & 08/11/2005 & 14704640 & 23.2	\\
1029: SMP~LMC~81 & 05$^{h}$35$^{m}$20$^{s}$.92 & $-$73$^{\rm o}$55$'$30$''$.1 & 10/10/2005 & 14704896 & 25	\\
1030: SMP~LMC~95 & 06$^{h}$01$^{m}$45$^{s}$.30 & $-$67$^{\rm o}$56$'$08$''$.0 & 08/11/2005 & 14705152 & 23.2	\\
1031: SMP~LMC~97 & 06$^{h}$10$^{m}$25$^{s}$.50 & $-$67$^{\rm o}$56$'$21$''$.0 & 08/11/2005 & 14705408 & 23.2	\\
1032: SMP~LMC~99 & 06$^{h}$18$^{m}$58$^{s}$.21 & $-$71$^{\rm o}$35$'$50$''$.7 & 10/15/2005 & 14705664 & 25	\\
1033: SMP~LMC~100 & 06$^{h}$22$^{m}$55$^{s}$.73 & $-$72$^{\rm o}$07$'$41$''$.4 & 11/14/2005 & 14699520 & 26	\\
1034: SMP~LMC~102 & 06$^{h}$29$^{m}$32$^{s}$.93 & $-$68$^{\rm o}$03$'$32$''$.9 & 08/11/2005 & 14699776 & 23.2	\\
1035: 
1036: 
1037: &&&&&\\
1038: SMP~SMC~2 & 00$^{h}$32$^{m}$38$^{s}$.81 & $-$71$^{\rm o}$41$'$58$''$.7 & 08/14/2005 & 14709248 & 23.2	\\
1039: SMP~SMC~5 & 00$^{h}$41$^{m}$21$^{s}$.67 & $-$72$^{\rm o}$45$'$18$''$.0 & 08/14/2005 & 14709504 & 23.2	\\
1040: SMP~SMC~8 & 00$^{h}$43$^{m}$25$^{s}$.17 & $-$72$^{\rm o}$38$'$18$''$.9 & 08/14/2005 & 14709760 & 23.2	\\
1041: SMP~SMC~9 & 00$^{h}$45$^{m}$20$^{s}$.66 & $-$73$^{\rm o}$24$'$10$''$.5 & 08/14/2005 & 14710016 & 23.2	\\
1042: SMP~SMC~13 & 00$^{h}$49$^{m}$51$^{s}$.71 & $-$73$^{\rm o}$44$'$21$''$.3 & 08/14/2005 & 14706176 & 23.2	\\
1043: SMP~SMC~14 & 00$^{h}$50$^{m}$34$^{s}$.99 & $-$73$^{\rm o}$42$'$57$''$.9 & 08/14/2005 & 14706432 & 23.2	\\
1044: SMP~SMC~15 & 00$^{h}$51$^{m}$07$^{s}$.45 & $-$73$^{\rm o}$57$'$37$''$.1 & 08/14/2005 & 14706688 & 23.2	\\
1045: SMP~SMC~16 & 00$^{h}$51$^{m}$27$^{s}$.08 & $-$72$^{\rm o}$26$'$11$''$.1 & 08/14/2005 & 14706944 & 23.2	\\
1046: SMP~SMC~17 & 00$^{h}$51$^{m}$56$^{s}$.44 & $-$71$^{\rm o}$24$'$44$''$.1 & 08/08/2005 & 14707200 & 23.2	\\
1047: SMP~SMC~18 & 00$^{h}$51$^{m}$57$^{s}$.97 & $-$73$^{\rm o}$20$'$30$''$.1 & 08/14/2005 & 14707456 & 23.2	\\
1048: SMP~SMC~19 & 00$^{h}$53$^{m}$11$^{s}$.14 & $-$72$^{\rm o}$45$'$07$''$.5 & 08/14/2005 & 14707712 & 23.2	\\
1049: SMP~SMC~20 & 00$^{h}$56$^{m}$05$^{s}$.39 & $-$70$^{\rm o}$19$'$24$''$.7 & 08/08/2005 & 14707968 & 23.2	\\
1050: SMP~SMC~23 & 00$^{h}$58$^{m}$42$^{s}$.14 & $-$72$^{\rm o}$56$'$59$''$.6 & 08/14/2005 & 14708224 & 23.2	\\
1051: SMP~SMC~25 & 00$^{h}$59$^{m}$40$^{s}$.51 & $-$71$^{\rm o}$38$'$15$''$.3 & 08/14/2005 & 14708480 & 23.2	\\
1052: SMP~SMC~26 & 01$^{h}$04$^{m}$17$^{s}$.81 & $-$73$^{\rm o}$21$'$51$''$.2 & 08/14/2005 & 14708736 & 23.2	\\
1053: SMP~SMC~27 & 01$^{h}$21$^{m}$10$^{s}$.67 & $-$73$^{\rm o}$14$'$35$''$.4 & 08/14/2005 & 14708992 & 23.2	\\
1054: 
1055: 
1056:    
1057: \enddata 
1058: \end{deluxetable} 
1059: 
1060: 
1061: 
1062: \begin{deluxetable}{lrcrll}
1063: 
1064: \tablecaption{Characteristics of the IRS spectra}
1065: 
1066: \tablehead{
1067: \colhead{Object} & 
1068: \colhead{T$_{\rm dust}$}& 
1069: \colhead{log F$_{\rm IR}$}& 
1070: \colhead{IRE} & 
1071: \colhead{Dust type} & 
1072: \colhead{Excitation} \\
1073: 
1074: \colhead{}& 
1075: \colhead{[K]}& 
1076: \colhead{[erg cm$^{-2}$ s$^{-1}$]} & 
1077: \colhead{}& 
1078: \colhead{}&
1079: \colhead{}\\
1080: }
1081: 
1082: \startdata
1083:   
1084:       SMP~LMC~4 & 190 & -11.7 & 2.65 & F & very high \\ 
1085:        SMP~LMC~9 & 140 & -11.7 & 0.86 & CRD & very high \\ 
1086:        SMP~LMC~10 & 90 & -12.5 & 0.10 & F & high \\ 
1087:        SMP~LMC~16 & 80 & -12.3 & 0.39 & F & very high \\ 
1088:        SMP~LMC~18 & 110 & -11.9 & 1.2 & F & intermediate \\ 
1089:        SMP~LMC~19 & 130 & -11.3 & 1.05 & CRD & very high \\ 
1090:        SMP~LMC~20 & 90& -12.3 & 0.24 & F & very high \\ 
1091:        SMP~LMC~21 & 130 & -11.1 & 20.96 & ORD & very high \\ 
1092:        SMP~LMC~25 & 160 & -10.7 & 1.52 & CRD & intermediate \\ 
1093:        SMP~LMC~27 & 140 & -11.9 & 1.75 & F & intermediate \\ 
1094:        SMP~LMC~34 & 130 & -11.4 & 1.40 & F & intermediate \\ 
1095:        SMP~LMC~45 & 120 & -11.9 & 0.30 & F & high \\ 
1096:        SMP~LMC~46 & 140 & -11.9 & 1.37 & CRD & high \\ 
1097:        SMP~LMC~48 & 170 & -10.8 & 1.43 & CRD & intermediate \\ 
1098:        SMP~LMC~66 & 140 & -11.6 & 0.96 & F & high \\ 
1099:        SMP~LMC~71 & 150 & -10.9 & 2.72 & CRD & very high \\ 
1100:        SMP~LMC~72 & 60 & -13.0 & 1.88 & F & very high \\ 
1101:        SMP~LMC~79 & 170 & -10.9 & 2.64 & CRD & high \\ 
1102:        SMP~LMC~80 & 110 & -13.1 & 0.05 & F & intermediate \\ 
1103:        SMP~LMC~81 & 130 & -10.6 & 2.79 & ORD & intermediate \\ 
1104:        SMP~LMC~95 & 110 & -12.1 & 0.68 & F & high \\ 
1105:        SMP~LMC~97 & 130 & -11.7 & 0.20 & F & very high \\ 
1106:        SMP~LMC~99 & 160 & -11.0 & 0.68 & CRD & very high \\ 
1107:       SMP~LMC~100 & 120 & -11.5 & 0.74 & CRD & very high \\ 
1108:       SMP~LMC~102 & 140 & -12.1 & 0.62 & F & high \\ 
1109:   
1110: 
1111: &&&&&\\
1112: 
1113: 
1114:        SMP~SMC~2 & 160 & -11.5 & 0.79 & CRD & very high \\ 
1115:        SMP~SMC~5 & 180 & -11.1 & $\dots$ & CRD & very high \\ 
1116:        SMP~SMC~8 & 160 & -11.7 & 0.49 & F & intermediate \\ 
1117:        SMP~SMC~9 & 120 & -12.0 & 1.10 & F & $\dots$ \\ 
1118:        SMP~SMC~13 & 190 & -11.3 & 0.60 & CRD & intermediate \\ 
1119:        SMP~SMC~14 & 150 & -11.6 & 0.91 & CRD & high \\ 
1120:        SMP~SMC~15 & 190 & -10.8 & 2.03 & CRD & intermediate \\ 
1121:        SMP~SMC~16 & 180 & -11.4 & 0.87 & CRD & low \\ 
1122:        SMP~SMC~17 & 160 & -11.1 & 1.19 & CRD & intermediate \\ 
1123:        SMP~SMC~18 & 170 & -10.8 & 2.37 & CRD & intermediate \\ 
1124:        SMP~SMC~19 & 150 & -11.3 & 1.50 & CRD & very high \\ 
1125:        SMP~SMC~20 & 250 & -11.0 & 1.22 & CRD & intermediate \\ 
1126:        SMP~SMC~23 & 150 & -12.2 & 0.35 & F & intermediate \\ 
1127:        SMP~SMC~25 & 130 & -11.5 & 1.86 & ORD & very high \\ 
1128:        SMP~SMC~26 & 130 & -12.1 & 0.76 & F & very high \\ 
1129:        SMP~SMC~27 & 180 & -11.4 & 0.47 & CRD & intermediate \\ 
1130: 
1131:    
1132: \enddata 
1133: \end{deluxetable} 
1134: %\end{document}
1135: 
1136: \begin{deluxetable}{lrrrrrrrr}
1137: \rotate
1138: \tablewidth{20truecm}
1139: \tablecaption{Summary of relations}
1140: 
1141: \tablehead{
1142: 
1143: \colhead{}&
1144: \colhead{}&
1145: \multicolumn{3}{c}{LMC}&
1146: \colhead{}&
1147: \multicolumn{3}{c}{SMC}\\
1148: \cline{3-5}
1149: \cline{7-9}
1150: \\
1151: 
1152: \colhead{}&
1153: \colhead{}&
1154: \colhead{F}&
1155: \colhead{CRD}&
1156: \colhead{ORD}&
1157: \colhead{}&
1158: \colhead{F}&
1159: \colhead{CRD}&
1160: \colhead{ORD}\\
1161: 
1162: }
1163: 
1164: \startdata
1165:   
1166: Sample size&	 &	14&	9&	2&	& 4&	11&	1\\
1167: 
1168: $<$C/H$>\times10^4$&	&	$(2.0\pm2.0)$& $(4.3\pm2.0)$& $(0.18\pm0.05)$& & $(4.2\pm2.7)$& $(4.6\pm2.0)$& $\dots$ \\
1169: 
1170: $<$N/H$>\times10^5$&	& $(8.4\pm0.1)$& $(7.6\pm4.1)$& $(12\pm15)$& & $(7.4\pm7.4)$& $(2.6\pm1.7)$& 8.3\\
1171: 
1172: $<$O/H$>\times10^4$&	&	$(1.8\pm1.0)$& $(2.4\pm1.0)$& $(1.3\pm0.7)$& & $(2.1\pm1.7)$& $(1.4\pm0.6)$& 0.4\\
1173: 
1174: $<$Ne/H$>\times10^5$&	&	$(2.7\pm1.5)$& $(4.1\pm1.9)$& $(1.8\pm0.4)$& & $(3.4\pm4.9)$& $(2.2\pm1.6)$& 0.9\\
1175: 
1176: 
1177: $<$M/M$_{\odot}>$&	& $(0.65\pm0.13)$& $(0.63\pm0.00)$& $\dots$&& 0.59& $(0.59\pm0.05)$& 0.82\\
1178: 
1179:   
1180:   
1181: \enddata 
1182: \end{deluxetable} 
1183: 
1184: \clearpage
1185: 
1186: \begin{figure}
1187: \figurenum{1}
1188: \plotone{f1.eps}
1189: \caption{Histogram of the [\ion{O}{3}] $\lambda$5007 flux of
1190: Magellanic Cloud PNe whose size is smaller
1191: than 2 $\arcsec$ as measured on the {\it HST} images. Outer histogram: complete sample. Gray histogram:
1192: the ROC sample (i.e., the GTO program by Houck). Black histogram: the PN sample presented in this paper.}
1193: \end{figure}
1194: 
1195: 
1196: 
1197: \begin{figure}
1198: \figurenum{2}
1199: \plotone{f2.eps}
1200: \caption{Continuum fitted (upper panel) and continuum-subtracted spectrum (lower panel)
1201: of SMP~SMC~15. Both nod positions are shown, in thin lines. The thick
1202: line in the upper plot identifies the blackbody fit used. Note that the fit is correctly
1203: below the broad dust features: the large silicon carbide feature at 11 $\mu$m, 
1204: the broad feature at 15--21 $\mu$m, and the broad feature peaking at 30 $\mu$m.
1205: Several nebular emission lines have
1206: been identified in the lower panel. The IRS spectral type of this PN is CRD (see text of $\S$2.3
1207: for spectral type definitions).
1208: }
1209: \end{figure}
1210: 
1211: \begin{figure}
1212: \figurenum{3}
1213: \plotone{f3.eps}
1214: \caption{Same as in Fig.~2, for SMP~LMC~100. The spectrum is dominated by the emission lines with 
1215: only a faint carbonaceous dust features at 6.2 $\mu$m. The IRS spectral type of this PN is CRD.
1216: }
1217: \end{figure}
1218: 
1219: 
1220: \begin{figure}
1221: \figurenum{4}
1222: \plotone{f4.eps}
1223: \caption{Same as in Fig.~2, for SMP~LMC~48. The broad plateau feature at 6--9 $\mu$m and the renmant of a broad
1224: amorphous SiC feature at 11 $\mu$m are clearly detected in this spectrum, together with a very strong 30 $\mu$m feature.
1225: The IRS spectral type of this PN is CRD.}
1226: \end{figure}
1227: 
1228: \begin{figure}
1229: \figurenum{5}
1230: \plotone{f5.eps}
1231: \caption{Same as in Fig.~2, for SMP~LMC~21. The oxygen-rich dust features are seen 
1232: at longer wavelengths. The IRS spectral type of this PN is ORD.}
1233: \end{figure}
1234: 
1235: 
1236: \clearpage
1237: \begin{figure}
1238: \epsscale{.8}
1239: \figurenum{6a}
1240: \plotone{f6a.eps}
1241: \caption{IRS spectra of F PNe.}
1242: \end{figure}
1243: \clearpage
1244: {\plotone{f6b.eps}}\\
1245: \centerline{Fig. 6b. --- Continued.}
1246: \clearpage
1247: {\plotone{f6c.eps}}\\
1248: \centerline{Fig. 6c. --- Contiunued.}
1249: \clearpage
1250: 
1251: 
1252: \begin{figure}
1253: \figurenum{7a}
1254: \plotone{f7a.eps}
1255: \caption{IRS spectra of CRD PNe. Left panel: entire spectra. Right panels: the 5--14 $\mu$m spectra, where the vertical
1256: lines correspond to the classical PAH features at 6.2, 7.7, 8.6, and 11.3 $\mu$m.}
1257: \end{figure}
1258: \clearpage
1259: {\plotone{f7b.eps}}\\
1260: \centerline{Fig. 7b. --- Continued.}
1261: \clearpage
1262: {\plotone{f7c.eps}}\\
1263: \centerline{Fig. 7c. --- Continued.}
1264: \clearpage
1265: 
1266: \begin{figure}
1267: \figurenum{8}
1268: \epsscale{1}
1269: \plotone{f8.eps}
1270: \caption{IRS spectra of ORD PNe. Left panel: entire spectra, where the vertical lines correspond
1271: to the crystalline silicate features at 23.5, 27.8, and 33.5 $\mu$m. Right panel: the 5--14 $\mu$m spectra.}
1272: \end{figure}
1273: 
1274: 
1275:  
1276:  \begin{figure}
1277: \figurenum{9}
1278: \includegraphics[bb = 100 100 600 600, angle=0]{f9.eps}
1279: \caption{Infrared luminosity versus carbon abundance. Left panel:
1280: dust type is coded with 
1281: triangles (F), diamonds (CRD), and squares (ORD). 
1282: Right panel: morphology is coded with circles (Round), diamonds (Elliptical), triangles
1283: (Bipolar Core), and squares (Bipolar, Quadrupolar, and Pointsymmetric).}
1284: \end{figure}
1285: 
1286: \begin{figure}
1287: \figurenum{10}
1288: \includegraphics[bb = 100 100 600 600, angle=0]{f10.eps}
1289: \caption{Carbon versus oxygen abundance of the gas of Magellanic Cloud PNe. Dust type (left panel)
1290: and morphology (right panel) are coded as in Fig.~9. The line represents C/O=1.}
1291: \end{figure}
1292: 
1293: \clearpage
1294: 
1295: \begin{figure}
1296: \figurenum{11}
1297: \includegraphics[bb = 100 100 600 600, angle=0]{f11.eps}
1298: \caption{The relation between T$_{\rm dust}$ and physical radius. Symbols for dust type (left panel)
1299: and morphology (right panel) are coded as in Fig.~9. The line is the best fit to the data. }
1300: \end{figure}
1301: 
1302: 
1303: 
1304: \begin{figure}
1305: \figurenum{12}
1306: \includegraphics[bb = 100 100 600 600, angle=0]{f12.eps}
1307: \caption{Dust temperature versus IR luminosity for dust type (left) and morphology (right).
1308: Symbols as in Fig. 9.}
1309: \end{figure}
1310: 
1311: \begin{figure}
1312: \figurenum{13}
1313: \includegraphics[bb = 100 100 600 600, angle=0]{f13.eps}
1314: \caption{Dust temperature versus electron density for dust type (left) 
1315: and morphology (right). Symbols as in Fig.~9.}
1316: \end{figure}
1317: 
1318: 
1319: 
1320: \begin{figure}
1321: \figurenum{14}
1322: \includegraphics[bb = 100 100 600 600, angle=0]{f14.eps}
1323: \caption{The loci of the central stars of a subsample of the studied PNe on the HR plane. Symbols of the observed PNe
1324: are as in Fig.~9.}
1325: \end{figure}
1326: 
1327: \begin{figure}
1328: \figurenum{15}
1329: \vspace*{-15mm}
1330: \plotone{f15.eps}
1331: \caption{Schematic evolutionary sequence of CRD PNe, where we plotted the order of apparent 
1332: evolution of the carbonaceous dust features, as in Figure
1333: 7a, b, and c, versus the photometric radii of the nebulae. Circles are 
1334: round PNe, diamonds are elliptical PNe, and triangles are bipolar core PNe.}
1335: \end{figure}
1336: 
1337: \end{document}
1338:     
1339: 	
1340: