0806.2318/ms.tex
1: \documentclass[12pt]{emulateapj}
2: %\usepackage{natbib}
3: 
4: \newcommand{\vdag}{(v)^\dagger}
5: \newcommand{\myemail}{as110@st-andrews.ac.uk}
6: 
7: %% You can insert a short comment on the title page using the command below.
8: 
9: \slugcomment{To appear in ApJL}
10: 
11: %% If you wish, you may supply running head information, although
12: %% this information may be modified by the editorial offices.
13: %% The left head contains a list of authors,
14: %% usually a maximum of three (otherwise use et al.).  The right
15: %% head is a modified title of up to roughly 44 characters.
16: %% Running heads will not print in the manuscript style.
17: 
18: \shorttitle{IRAS04325C}
19: \shortauthors{Scholz et al.}
20: 
21: %% This is the end of the preamble.  Indicate the beginning of the
22: %% paper itself with \begin{document}.
23: 
24: \begin{document}
25: \bibliographystyle{apj}
26: 
27: %% LaTeX will automatically break titles if they run longer than
28: %% one line. However, you may use \\ to force a line break if
29: %% you desire.
30: 
31: \title{IRAS04325+2402C: A very low mass object with an edge-on disk}
32: 
33: %% Use \author, \affil, and the \and command to format
34: %% author and affiliation information.
35: %% Note that \email has replaced the old \authoremail command
36: %% from AASTeX v4.0. You can use \email to mark an email address
37: %% anywhere in the paper, not just in the front matter.
38: %% As in the title, use \\ to force line breaks.
39: 
40: \author{Alexander Scholz\altaffilmark{1}, Ray Jayawardhana\altaffilmark{2}, Kenneth Wood\altaffilmark{1}, 
41: David Lafreni\`ere\altaffilmark{2,5}, Katharina Schreyer\altaffilmark{3}, Ren\'e Doyon\altaffilmark{4}}
42: 
43: \email{as110@st-andrews.ac.uk}
44: 
45: \altaffiltext{1}{SUPA, School of Physics \& Astronomy, University of St. Andrews, North Haugh, St. Andrews, 
46: KY16 9SS, United Kingdom}
47: %\author{Ray Jayawardhana, David Lafreni\`ere}
48: \altaffiltext{2}{Department of Astronomy \& Astrophysics, University of Toronto, 50 St. George Street, Toronto, 
49: ON M5S 3H4, Canada}
50: %\author{Katharina Schreyer}
51: \altaffiltext{3}{Astrophysikalisches Institut und Universit{\"a}ts-Sternwarte Jena, Schillerg{\"a}sschen 2-3,
52: D-07745 Jena, Germany}
53: %\author{Ren\'e Doyon}
54: \altaffiltext{4}{D\'epartement de Physique and Observatoire du Mont M\'egantic, Universit\'e de Montr\'eal, C.P. 6128, 
55: Succursale Centre-Ville, Montreal, QC H3C 3J7, Canada}
56: 
57: \altaffiltext{5}{Visiting Astronomer at the Infrared Telescope Facility, 
58: which is operated by the University of Hawaii under Cooperative Agreement no. NNX08AE38A 
59: with the National Aeronautics and Space Administration, Science Mission Directorate, 
60: Planetary Astronomy Program.}
61: 
62: \begin{abstract}
63: IRAS04325+2402C is a low luminosity object located near a protostar in Taurus.
64: We present new spatially-resolved mm observations, near-infrared spectroscopy, and 
65: Spitzer photometry that improve the constraints on the nature of this source. The 
66: object is clearly detected in our 1.3\,mm interferometry map, allowing us to 
67: estimate the mass in a localized disk$+$envelope around it to be in the range of 
68: 0.001 to $0.01\,M_{\odot}$. Thus IRAS04325C is unlikely to accrete significantly more 
69: mass. The near-infrared spectrum cannot be explained with an extincted photosphere 
70: alone, but is consistent with a 0.03-0.1$\,M_{\odot}$ central source plus moderate 
71: veiling, seen in scattered light, confirming the edge-on nature of the disk. Based on 
72: K-band flux and spectral slope we conclude that a central object mass $\gtrsim 0.1\,M_{\odot}$ 
73: is unlikely. Our comparison of the full spectral energy distribution, including 
74: new {\it Spitzer} photometry, with radiative transfer models confirms the high inclination 
75: of the disk ($\gtrsim 80\arcdeg$), the very low mass of the central source, and the small 
76: amount of circumstellar material. IRAS04325C is one of the lowest mass objects with a 
77: resolved edge-on disk known to date, possibly a young brown dwarf, and a likely wide 
78: companion to a more massive star. With these combined properties, it represents a 
79: unique case to study the formation and early evolution of very low mass objects.
80: \end{abstract}
81: 
82: \keywords{stars: circumstellar matter, formation, low-mass, brown dwarfs -- planetary systems}
83: 
84: 
85: \section{Introduction}
86: \label{intro}
87: 
88: The origin of brown dwarfs has been one of the most intensely debated subjects in cool stars
89: research over the past decade. As of today, there is no consensus on the dominant mechanisms 
90: that determine the low-mass end of the stellar initial mass function. Recent modeling has 
91: provided predictions for observable properties in the very low mass (VLM) regime, which in 
92: principle would permit distinguishing between the proposed scenarios. In practice, however, 
93: it has turned out to be challenging to place firm limits on the formation theory for VLM 
94: sources \citep[see reviews by][]{2007prpl.conf..459W,2007prpl.conf..443L}.
95: 
96: Of particular relevance for evaluating formation scenarios are the global properties of 
97: disks and envelopes. The combination of high-resolution imaging and analysis of the spectral 
98: energy distribution (SED) in the infrared/mm regime is generally acknowledged as the optimum 
99: source of information for disk/envelope parameters of young stellar sources. As of today, 
100: constraints on radii and masses of brown dwarf disks are sparse: Recent deep mm observations 
101: provided first limits on disk masses ($<0.001-0.003\,M_{\odot}$) and outer radii
102: \citep[10-100\,AU;][]{2006ApJ...645.1498S,2003ApJ...593L..57K}. Only one substellar disk has been 
103: resolved with high-resolution imaging \citep{2007ApJ...666.1219L}, with a derived radius
104: of 20-40\,AU.
105: 
106: In this context, the young source IRAS04325+2402C (hereafter: IRAS04325C) in the Taurus
107: star forming complex may be of particular interest. HST/NICMOS images of the Class I source 
108: IRAS04325+2402 \citep{1995ApJS..101..117K}, also known as L1535\,IRS, revealed  IRAS04325C as a 
109: faint companion, at a separation of $8\farcs1$ \citep{1999AJ....118.1784H}\footnote{Note that
110: the primary itself is likely to be a subarcsecond binary, see \citet{1999AJ....118.1784H}}. The object 
111: is resolved `into a double-lobed emission separated by a dark lane' \citep{1999AJ....118.1784H}, 
112: resembling the images of edge-on T Tauri disks. Models including an edge-on disk and an envelope 
113: are able to reproduce the appearance of the images. From the near-infrared (NIR) photometry, the mass 
114: of this possible Class I object was estimated to be between 0.02 and 0.06\,$M_{\odot}$, with an upper 
115: limit of 0.25\,$M_{\odot}$. Based on statistical arguments, \citet{2004A&A...427..651D} find that the 
116: system is likely to be a physically bound binary  (probability 98\%). 
117: 
118: Thus, IRAS04325C is a potential benchmark object to study substellar formation. In this Letter, 
119: we present new millimeter interferometry and NIR spectroscopy as well as an analysis of 
120: the resolved SED for this source, including new {\it Spitzer} photometry, with the aim of 
121: improving the constraints on the properties of the system, particularly the mass of the central 
122: object and the disk$+$envelope.
123: 
124: \section{New constraints from resolved observations}
125: \label{obs}
126: 
127: \subsection{Millimeter interferometry}
128: \label{mm}
129: 
130: Using the IRAM Plateau de Bure Interferometer \citep{1992A&A...262..624G}, we obtained a 1.3\,mm 
131: (226.50 GHz) continuum map centered on IRAS04325C, observed on 27 December 2007 (configuration C, 
132: rms 0.3\,mJy/beam, HPBW $1.55"\times 1.0"$). The final map of 128$\times$128 square pixels with 
133: $0\farcs53$ pixel size was produced by Fourier transforming the calibrated visibilities 
134: using natural weighting using the Grenoble Software environment GAG. 
135: 
136: \begin{figure}
137: \center
138: \includegraphics[angle=-90,width=6cm]{f1.eps}
139: \caption{Interferometry map at 1.3\,mm wavelength, obtained with the IRAM Plateau de Bure Interferometer. 
140: The map is centered on our target IRAS04325C; the primary source is seen at (+2",-8"). The inlet in the 
141: lower left corner shows the beam size. The contours are 10\% to 90\% of the peak value in steps of 10\%. 
142: \label{f1}}
143: \end{figure}
144: 
145: The 1.3\,mm image (Fig. \ref{f1}) shows two sources, which are identified with the primary 
146: (south) and companion (north) object in the HST images from \citet{1999AJ....118.1784H}. We 
147: estimate fluxes of $10\pm ^3_1$\,mJy for both sources by integrating the pixel intensities. 
148: The 1.3\,mm flux is  attributed to thermal emission from cold dust. Since the peaks in the
149: map are strongly centered at the positions of the infrared sources, the emission most likely 
150: originates from the disk and the localized envelope with a diameter $\lesssim 300\,AU$. Assuming 
151: that the dust is optically thin, we can obtain an estimate of the total dust mass $M_\mathrm{dust}$ 
152: from the mm flux $S_\nu$:
153: \begin{equation}
154: M_\mathrm{dust} = \frac{S_\nu D^2}{B_\nu(T_\mathrm{D})\kappa_\nu}
155: \end{equation}
156: We assume a distance $D$ of 148\,pc for the Taurus star forming region 
157: \citep{2007ApJ...671..546L}, a gas to dust ratio of 100, and a plausible range of values for 
158: dust temperature $T_\mathrm{D}$ (10-25\,K) and dust opacity $\kappa_\nu$ (1-3\,cm$^2$g$^{-1}$ at 1.3\,mm), 
159: as commonly suggested in the literature \cite[see][and references therein]{2006ApJ...645.1498S}. Using 
160: these parameters, we obtain a total mass (dust and gas) of 1-10\,$M_{\mathrm{Jup}}$ and a most probable 
161: value of 3\,$M_{\mathrm{Jup}}$ for each of the two sources. This value is comparable to the highest 
162: disk mass determined for a Class II brown dwarf in Taurus \citep{2006ApJ...645.1498S}.
163: 
164: Single-dish measurements centered on IRAS04325 are available in the literature and can be used 
165: to put limits on the mass of a possible common envelope surrounding both the primary and the 
166: secondary. Published fluxes are $\sim 180$\,mJy at 850\,$\mu m$ 
167: \citep[15" beam,][]{2000ApJ...534..880H,2005ApJ...631.1134A}, 110\,mJy at 1.3\,mm 
168: \citep[11",][]{2001A&A...365..440M}, $<9.2$\,mJy at 3\,mm \citep[$\sim 11"$,][]{1996ApJ...466..317O}.
169: Using the same parameters as above and $\kappa_\nu \propto \lambda^{-\beta}$ with $\beta\sim1.5$ 
170: \citep{2007prpl.conf..767N} yields corresponding masses of 10-30\,$M_{\mathrm{Jup}}$ of distributed 
171: dust and gas in an area with diameter $\sim 2000$\,AU around the primary IRAS source, including the 
172: companion. Fluxes integrated over larger areas \citep[500\,mJy for 60" diameter,][]{2001A&A...365..440M} 
173: clearly indicate the presence of large-scale mm background in that area. Given that the primary in the
174: system is probably more massive than the companion (see Sect. \ref{nature}), IRAS04325C is unlikely to
175: be the dominant center of accretion in that area. Thus, the total mass reservoir for IRAS04325C is 
176: 1-10$\,M_{\mathrm{Jup}}$ in the localized disk/envelope plus presumably a small fraction of at 
177: most 30\,$M_{\mathrm{Jup}}$ in a possible common envelope. 
178: 
179: \subsection{Near-infrared spectroscopy}
180: \label{spec}
181: 
182: We obtained a NIR (1-2.5\,$\mu m$) low-resolution ($R\sim 100$) spectrum for IRAS04325C using SpeX at IRTF
183: \citep{2003PASP..115..362R}. Reduction, background correction, and extraction was carried 
184: out using SpeXtools \citep{2004PASP..116..362C}. We corrected for instrument response and telluric 
185: features using a spectrum of the A0 star HD25175 observed immediately before the target. The final
186: signal-to-noise ratio ranges from 15 in H- to 30 in K-band. Our spectrum rises steeply from J- to 
187: K-band and is lacking obvious photospheric features. The spectrum is shown in Fig. \ref{f2} (solid line) 
188: scaled to the published K-band flux \citep[14.0-14.6\,mag,][]{1999AJ....118.1784H,2008AJ....135.2496C}. In 
189: the following, we will use the shape of the spectrum and the K-band flux to derive a constraint on 
190: the mass of the central source.
191: 
192: \begin{figure}
193: \center
194: \includegraphics[angle=-90,width=8cm]{f2.ps}
195: \caption{IRTF/SpeX spectrum for IRAS04325C (solid line) in comparison with models (dotted lines). All
196: spectra are scaled to the K-band flux. The models assume $T_{\mathrm{eff}}=3000$\,K and typical veiling. 
197: The three model spectra in the plot are calculated from three scattering models 
198: \citep[MRN2, 12, 15 in][]{1993ApJ...414..773K}
199: \label{f2}}
200: \end{figure}
201: 
202: We compare the observed spectrum with models based on the photospheric templates STARdusty2000
203: and BDdusty2000 \citep{2001ApJ...556..357A}. In a first attempt, we calculated
204: a series of models assuming pure extinction, thus neglecting the edge-on nature 
205: \citep[NIR extinction law from][]{1990ARA&A..28...37M}. While it is possible to reproduce 
206: the slope of the spectrum using such models, the required combinations of $T_{\mathrm{eff}}$ and 
207: $A_V$ imply photospheric K-band fluxes that are more than one order of magnitude larger than the 
208: observed value. Thus, pure extinction cannot account for the NIR SED, confirming that the object 
209: is seen in scattered light through an edge-on disk.
210: 
211: Thus, instead of extinction, we applied scattering models from \citet{1993ApJ...414..773K}, which have
212: been confirmed to fit the colours of embedded young stars in Taurus. For a range of scattering properties
213: and edge-on geometry, \citet{1993ApJ...414..773K} obtain 2.0-4.5\,mag difference between observed and 
214: photospheric K-band flux, in line with the properties of many known edge-on disks. This implies for 
215: our target $M_K=3.8\ldots6.9$\,mag. Comparison with the 1\,Myr track by \citet{1998A&A...337..403B} yields $T_{\mathrm{eff}}=2700\ldots3200$\,K and $M=0.03\ldots0.16\,M_{\odot}$.
216: 
217: \begin{figure}
218: \center
219: \includegraphics[angle=0,width=8cm]{f3.ps}
220: \caption{Spectral energy distribution for IRAS04325C in comparison with Monte Carlo radiative transfer 
221: models. Triangles and diamonds are the observed fluxes; Spitzer datapoints at 3.6-8$\,\mu m$ are treated
222: as upper limits. Solid line: model with ISM grains, dashed line: model with larger grains (see text),
223: dash-dotted line: model with face-on geometry, dotted line: input photosphere. (All models include 
224: extinction of $A_V=10$.)
225: \label{f3}}
226: \end{figure}
227: 
228: Photospheric spectra for this range of temperatures exhibit deep $H_2O$ absorption features, 
229: which are not seen in our spectrum. The most likely reason for their absence is veiling, 
230: as frequently observed for young stellar sources \citep{1996AJ....112.2184G}. We use
231: the typical veiling for Class-II sources of $r_K=1.76$ and $r_J=0.97$ \citep{1999A&A...352..517F}.
232: If the K-band flux includes that amount of veiling, the upper limits for temperature and mass are
233: reduced to $T_{\mathrm{eff}}=3000$\,K and $M=0.1\,M_{\odot}$.
234: 
235: For both scattering and veiling, we fit the values for the photometric bands with a smooth 
236: function to obtain the wavelength dependence. Veiling is added first, assuming that all excess flux 
237: originates in regions close to the central object and is thus affected by scattering. Under these 
238: assumptions, we can fit the observed SED with $T_\mathrm{eff}\sim 3000$\,K, corresponding to  
239: a mass of $\sim 0.1\,M_{\odot}$ at 1.0\,Myr. In Fig. \ref{f2} we show spectra for three 
240: plausible scattering models \citep[MRN2, 12, 15 in][]{1993ApJ...414..773K}, all with 
241: $T_\mathrm{eff}=3000$\,K and veiling, as described above. Stronger veiling, as usually seen for
242: Class I sources \citep{2005AJ....130.1145D}, would further reduce the photospheric contribution
243: to the K-band flux and allow for deeper $H_2O$ features, thus lower $T_\mathrm{eff}$ and mass. 
244: 
245: Interpreting the NIR SED for IRAS04325C is challenging and subject to large uncertainties.
246: We have shown that under reasonable assumptions the observational constraints are well-explained 
247: by veiled emission from a central object with 0.03-0.1$\,M_{\odot}$ seen through an edge-on disk. 
248: Higher masses cannot be strictly ruled out, but would require unusual scattering
249: characteristics. 
250: 
251: \subsection{Spitzer photometry}
252: \label{spitzer}
253: 
254: IRAS04325 has been observed with Spitzer IRAC/MIPS as part of the Taurus Spitzer Legacy Project (PI: D. Padgett). 
255: The spatial resolution of MIPS is not sufficient to properly resolve the system; the combined fluxes for the source, 
256: as given in the Taurus-1 source catalogue, are 1.9\,Jy at 24\,$\mu m$ and $4.7\pm0.9$\,Jy at 70\,$\mu m$. For comparison:
257: The IRAS mid-infrared fluxes for this object are 2.1\,Jy at 25\,$\mu m$ and 12.86\,Jy at 60\,$\mu m$, again
258: indicating that long-wavelength data with large apertures include significant extended background emission
259: (see Sec. \ref{mm}).
260: 
261: In the IRAC images, the object is clearly not point-like and shows two emission peaks coinciding with
262: the positions of the two components, superimposed onto a bright nebulosity. We carried out resolved
263: photometry at 3-8\,$\mu m$ by fitting the PSF in apertures with $3\farcs6$ radius centered on the peaks 
264: of the two components. We estimate the fluxes of the companion to be 4.1, 3.5, 2.6, 1.8\,mJy at wavelengths 
265: 3.6, 4.5, 5.8, and 8.0\,$\mu m$ (uncertainty $\pm 20$\%). Owing to the difficulties in disentangling the 
266: emission from the sources and from the underlying nebulosity, the true Spitzer fluxes of IRAS04325C are 
267: likely to be significantly smaller. Comparing with the L-band flux published by \citet{2008AJ....135.2496C} 
268: indicates that the background contamination may be as much as $\sim 70$\%. The fluxes of the primary source 
269: in the system are about 6, 7, 10, 11 times higher for IRAC channels 1-4, respectively.
270: 
271: \section{Modeling the SED of IRAS04325C}
272: \label{model}
273: 
274: We compiled the SED for IRAS04325C from our new observations and the 
275: literature. In the NIR, we aim to reproduce the K- and L-band fluxes from
276: \citet{2008AJ....135.2496C}. The Spitzer fluxes are considered to be upper limits, as outlined 
277: in Sec. \ref{spitzer}. In Fig. \ref{f3} we compare the observed SED with three model SEDs from a 
278: Monte Carlo radiation transfer code, which has been used successfully to model the SEDs 
279: of T Tauri stars \citep{2002ApJ...564..887W} and brown dwarfs \citep{2006ApJ...645.1498S}. The models 
280: use NextGen model atmospheres for the photosphere \citep{2001ApJ...556..357A} and include dust 
281: destruction close to the central object. Accretion rates are assumed to be negligible for the heating 
282: of the disk. For more details on the model ingredients, see \citet{2006ApJ...645.1498S}. 
283: In all three models, we fix the disk radius at 30\,AU, as given by \citet{1999AJ....118.1784H} 
284: based on the analysis of the HST images, the extinction at $A_V=10$, and the parameters of the 
285: central object at values consistent with our spectroscopy ($T_\mathrm{eff}=2600$\,K, $M=0.07\,M_{\odot}$).
286: In addition, the model contains an infalling envelope with a bipolar cavity. 
287: 
288: The models shown with solid and dashed line assume an inclination of $i=80\arcdeg$ and two different
289: grain size distributions: small ISM-type grains (solid line) and larger grains (dashed line), mimicking 
290: the effect of dust coagulation. In the latter case, the grain size distribution is a power law with 
291: an exponential decay for particles with sizes above 50$\,\mu m$ and a formal maximum grain size of 1\,mm,
292: see \citet{2002ApJ...564..887W}. Both models are able to provide a good match to the available datapoints. 
293: To fit the mm datapoint we need a total circumstellar mass of $2.2\times 10^{-3}\,M_{\odot}$ for ISM grains 
294: and $4.6\times 10^{-4}\,M_{\odot}$ for the larger grains. In the second case, the value for the disk/envelope
295: mass is lower than the constraint given in Sec. \ref{mm} due to the higher dust opacity. Note that the two 
296: models are highly discrepant in their prediction for the fluxes at  $5-20\,\mu m$ due to the different 
297: scattering albedo of the two grain types.
298: 
299: For illustration purposes, we also provide the SED for a face-on geometry in dash-dotted lines.
300: Clearly, the model is not in agreement with the observations. No combination of $T_\mathrm{eff}$ and
301: $A_V$ is able to fit both the observed fluxes and the near-infrared spectral slope, as discussed in
302: Sec. \ref{spec}.
303: 
304: In summary, the modeling clearly confirms the plausibility of the results obtained in earlier 
305: sections. The observed SED can be reproduced assuming that the central source is a very low mass star
306: or brown dwarf, the inclination is high (i.e. the object is seen mostly in scattered light), and the
307: circumstellar mass is negligible compared with the object mass. The models also demonstrate 
308: that more complete wavelength coverage is needed for further analysis of the properties 
309: of the system.
310: 
311: \section{The nature of IRAS04325C}
312: \label{nature}
313: 
314: We present new observational constraints on the nature of the potential young brown dwarf IRAS04325C,
315: a $8\farcs1$ companion to a well-known IRAS source discovered by \citet{1999AJ....118.1784H} in 
316: HST/NICMOS images. From our NIR spectrum for IRAS04325C, we estimate that the central source 
317: likely has a mass between 0.03 and 0.1$\,M_{\odot}$. The available constraints on the infrared SED 
318: are inconsistent with a reddened photospheric spectrum, but they can be understood by assuming that 
319: the object is seen in scattered light. This is independent confirmation for the edge-on geometry of 
320: the disk, as seen in the HST images. IRAS04325C is only the second case of an edge-on system in the 
321: mass range $\lesssim 0.1\,M_{\odot}$ that has been resolved with high-resolution imaging. 
322: 
323: We detect the source with 1.3\,mm interferometry at a flux level of 10\,mJy, corresponding to a mass
324: of 1-10\,$M_{\mathrm{Jup}}$ in a localized disk/envelope with $R\lesssim 300$\,AU. Since the accretion
325: reservoir in a larger-scale envelope is probably insignificant, the central source is unlikely to
326: gain more mass than 0.01\,$M_{\odot}$ due to further accretion. Thus, IRAS04325C will probably become 
327: either a very low mass star close to the substellar boundary or a brown dwarf. 
328: 
329: The evolutionary stage of IRAS04325C is not well-determined based on the current data.
330: The primary component of the system has been classified as a Class I source based on the mid-infrared 
331: spectral slope \citep[$\alpha(2-25)=0.79$,][]{2004A&A...427..651D} and the NIR colours 
332: \citep{1992ApJ...393..278L}. It is also assumed to drive a CO outflow \citep[e.g.][]{1998ApJ...502..315H}, 
333: another characteristic for a protostellar source. The companion is clearly embedded in the same cloud core 
334: and likely physically bound to the primary, thus the two sources can be assumed to be coeval. This would 
335: make IRAS04325C the lowest mass Class I source known to date. The classification of the primary source, 
336: however, is partly based on unresolved data, including light from the companion and the surrounding 
337: nebulosity. Furthermore, the disk of the primary is probably seen close to edge-on 
338: \citep[$i\sim 80\arcdeg$,][]{2008ApJS..176..184F}, which would suppress the NIR flux and might lead 
339: to its misidentification as Class I source \citep[see][]{2007prpl.conf..117W}. We estimate a total mass 
340: of a few Jupiter masses in a localised disk$+$envelope for the primary source, which seems to be too low 
341: for a Class I object \citep{2005ApJ...631.1134A}. Thus, the Class I nature of the source is questionable. 
342: 
343: The central source of IRAS04325C has probably significantly less mass than the primary component in the system.
344: In all near/mid-infrared-bands, IRAS04325C is about an order of magnitude fainter than the primary 
345: source, indicating a mass ratio of $\sim 0.15$. The luminosity of the primary has been estimated to be 
346: 0.7-0.9$L_{\odot}$ \citep{1993ApJ...414..773K,2008ApJS..176..184F}, implying a mass of 
347: $\sim 0.7\,M_{\odot}$. IRAS04325C also has a much smaller disk than the primary object, as seen 
348: in the HST images.
349: 
350: With its combined properties -- very low mass central object, edge-on disk, wide companion to a more massive
351: star -- IRAS04325C is a prime target for follow-up studies and presents us with a unique opportunity 
352: to study the early evolutionary stages in a critical mass range.
353: 
354: \acknowledgments
355: This work is based on observations carried out with the IRAM Plateau de Bure Interferometer. IRAM 
356: is supported by INSU/CNRS (France), MPG (Germany) and IGN (Spain). We would like to thank the staff at 
357: IRAM, in particular Jan-Martin Winters and Vincent Pietu, for support with observations and data reduction.
358: The help provided by John Rayner during the SpeX observations is appreciated. Parts of this study 
359: are based on archival data obtained with the Spitzer Space Telescope, which is operated by the Jet 
360: Propulsion Laboratory, California Institute of Technology under a contract with NASA. 
361: 
362: \bibliography{aleksbib}
363: 
364: \end{document}
365: 
366: 
367: