1: \documentclass[manuscript]{aastex}
2: %\documentclass[12pt,preprint]{aastex}
3: %\documentclass[preprint2]{aastex}
4:
5: \usepackage{emulateapj5}
6:
7: \newcommand{\Mj}{\,$M_{\rm Jup}$}
8: \newcommand{\Ms}{\,$M_{\odot}$}
9:
10: \slugcomment{Accepted for publication in the ApJ}
11:
12: \shorttitle{A Methane Isolated Planetary Mass Object in Orion}
13: \shortauthors{Zapatero Osorio et al.}
14:
15: \begin{document}
16:
17: \title{A Methane Isolated Planetary Mass Object in Orion}
18:
19: \author{M.\,R$.$ Zapatero Osorio\altaffilmark{1},
20: V.\,J.\,S$.$ B\'ejar\altaffilmark{2},
21: E.\,L$.$ Mart\'\i n\altaffilmark{3},
22: R$.$ Rebolo\altaffilmark{2,4},
23: D$.$ Barrado y Navascu\'es\altaffilmark{1},
24: R$.$ Mundt\altaffilmark{5},
25: J$.$ Eisl\"offel\altaffilmark{6},
26: and J.\,A$.$ Caballero\altaffilmark{2}}
27:
28: \altaffiltext{1}{LAEFF-INTA, P.\,O$.$ Box 50727, E-28080 Madrid, Spain;
29: mosorio@laeff.esa.es, barrado@laeff.esa.es}
30: \altaffiltext{2}{Instituto de Astrof\'\i sica de Canarias, E-38205
31: La Laguna, Tenerife, Spain; vbejar@ll.iac.es,
32: rrl@ll.iac.es, zvezda@ll.iac.es}
33: \altaffiltext{3}{Institute of Astronomy, Univ. of Hawaii at Manoa, 2680
34: Woodlawn Drive, Honolulu, HI 96822 USA;
35: ege@ifa.hawaii.edu}
36: \altaffiltext{4}{Consejo Superior de Investigaciones Cient\'\i ficas,
37: Madrid, Spain}
38: \altaffiltext{5}{Max-Planck-Institut f\"ur Astronomie, K\"onigstuhl 17,
39: D-69117 Heidelberg, Germany; mundt@mpia-hd.mpg.de}
40: \altaffiltext{6}{Th\"uringer Landessternwarte Tautenburg, Sternwarte 5,
41: D-07778 Tautenburg, Germany; jochen@tls-tautenburg.de}
42:
43:
44: \begin{abstract}
45: We report on the discovery of a free-floating methane dwarf toward the
46: direction of the young star cluster $\sigma$\,Orionis. Based on the
47: object's far-red optical and near-infrared photometry and
48: spectroscopy, we conclude that it is a possible member of this
49: association. We have named it as S\,Ori\,J053810.1--023626 (S\,Ori\,70
50: is the abridged name). If it is a true member of $\sigma$\,Orionis,
51: the comparison of the photometric and spectroscopic properties of
52: S\,Ori\,70 with state-of-the-art evolutionary models yields a mass of
53: 3\,$^{+5}_{-1}$ Jupiter mass for ages between 1\,Myr and 8\,Myr. The
54: presence of such a low-mass object in our small search area
55: (55.4\,arcmin$^2$) would indicate a rising substellar initial mass
56: function in the $\sigma$ Orionis cluster even for planetary masses.
57: \end{abstract}
58:
59:
60: \keywords{stars: individual (S\,Ori\,70) ---
61: stars: low-mass, brown dwarfs --- stars: pre-main sequence --- open
62: clusters and associations: individual ($\sigma$\,Orionis)}
63:
64:
65: \section{Introduction}
66:
67: Since the discovery of brown dwarfs (i.e., objects unable to burn
68: hydrogen stably in their interiors and with masses below 72\Mj; Kumar
69: \cite{kumar63}; Chabrier et al$.$ \cite{chabrier00}) both in the field
70: and in young open clusters (see Basri \cite{basri00} for a review),
71: many questions remain unsolved. A very important one is the minimum
72: mass for the formation of very low mass objects in isolation, which
73: would represent the bottom end of the initial mass function (IMF) for
74: free-floating objects. B\'ejar et al$.$ \cite{bejar01} have
75: scrutinized the substellar population of the young, nearby $\sigma$
76: Orionis cluster, and derived a rising IMF
77: (d$N$/d$M$\,$\sim$\,$M^{-0.8\pm0.4}$), which is complete for brown
78: dwarf masses in the range of 72--13\Mj. Other authors find similar
79: substellar IMFs (e.g., Luhman et al$.$ \cite{luhman00}). Very recent
80: photometric and spectroscopic searches (Lucas \& Roche \cite{lucas00};
81: Zapatero Osorio et al$.$ \cite{osorio00}) suggest that the IMF extends
82: further below the deuterium burning mass threshold at around
83: 13\Mj~(Saumon et al$.$ \cite{saumon96}; Burrows et al$.$
84: \cite{burrows97}). We will refer to this mass regime as the
85: ``planetary mass'' domain. The least-massive objects so far identified
86: in young stellar clusters of Orion have masses around 5--10\Mj~(Lucas
87: et al$.$ \cite{lucas01}; Mart\'\i n et al$.$ \cite{martin01}), and
88: cover the full range of the spectral type L. For the formation
89: scenarios of such objects it would be very important to see down to
90: which masses they can be found in isolation.
91:
92: Here we report on the discovery of a methane dwarf toward the
93: direction of Orion. In this paper we present evidence for its
94: membership in the $\sigma$\,Orionis star cluster, which implies that
95: this object is likely the least massive planetary mass body imaged to
96: date outside the solar system. Cluster membership is discussed in
97: Section \ref{membership}, and mass determination based on
98: state-of-the-art evolutionary models and further discussion are
99: presented in Section \ref{mass}.
100:
101: \section{Observations \label{obs}}
102:
103: Our candidate was selected from a $J$, $H$ near-infrared survey, in
104: which the south-western region of the young $\sigma$\,Orionis cluster
105: was targeted down to 3-$\sigma$ detection limit of $J, H \sim$
106: 21\,mag. An area of 55.4\,arcmin$^2$ was covered with the
107: near-infrared camera INGRID mounted at the Cassegrain focus of the
108: 4.2-m William Herschel Telescope (WHT; at Roque de los Muchachos
109: Observatory). This camera is equipped with a 1024$\times$1024 Hawaii
110: detector, which provides a pixel size of 0\arcsec.242 projected onto
111: the sky. The observations were carried out on 2000 November 5--6;
112: these nights were photometric, but the seeing was quite poor at around
113: 1\arcsec.8. The total integration time was 3240\,s in each of the $J$
114: and $H$ filters. A nine-position dither pattern, repeated three times,
115: was used to obtain the images; each image consists of 4 co-added
116: exposures of 30\,s. Standard data reduction, which included dark
117: subtraction and flat-fielding, was done under the {\sc
118: iraf\footnote{IRAF is distributed by National Optical Astronomy
119: Observatory, which is operated by the Association of Universities for
120: Research in Astronomy, Inc., under contract with the National Science
121: Foundation.}} environment. Instrumental magnitudes were converted
122: into observed magnitudes using short integration observations of the
123: INGRID fields obtained with the near-infrared camera available at the
124: 1.5-m Carlos S\'anchez Telescope (Teide Observatory) on the
125: photometric nights of 2002 April 2--3. UKIRT standard stars taken from
126: the work of Hunt et al$.$ \cite{hunt98} were observed throughout this
127: night.
128:
129: Among various photometric candidates for possible membership in
130: $\sigma$\,Orionis, one showed a rather blue $J-H$ color of about
131: --0.1\,mag and $J$\,=\,20.28\,mag. We inspected previous $I$- and
132: $z$-band images obtained with the Low Resolution Imaging Spectrometer
133: (LRIS; Oke et al$.$ \cite{oke95}) at the 10-m Keck II telescope on
134: 1998 December 23 to identify this INGRID candidate -- the INGRID
135: survey mostly overlapped with the LRIS optical images. It appeared at
136: the detection limit with $I$\,=\,25.0\,mag and
137: $I-z$\,=\,2.2\,mag. LRIS is equipped with a Tek 2048\,$\times$\,2048
138: CCD detector, which provides an imaging scale of
139: 0\arcsec.215\,pix$^{-1}$. Our optical LRIS survey covered an area of
140: 218.4\,arcmin$^2$ in the $\sigma$\,Orionis cluster. Individual
141: exposures of 5\,min integration time were taken in each filter; the
142: average seeing was 0\arcsec.7--0\arcsec.8. $I$-band magnitudes were
143: calibrated with respect to previously known $\sigma$\,Orionis brown
144: dwarfs, like S\,Ori\,47 (Zapatero Osorio et al$.$
145: \cite{osorio99a}). The $z$-band data were calibrated as in Zapatero
146: Osorio et al$.$ \cite{osorio99b}; this is not an absolute calibration
147: and is relative to the color $I-z$, which is supposed to be 0.0 for
148: A0-type stars. Further details on this LRIS search will be given
149: elsewhere (B\'ejar et al$.$ \cite{bejar02}). Based on the $IzJH$
150: colors, our candidate turns out to be an extremely red object. We have
151: named it as S\,Ori\,70; IAU nomenclature and coordinates (accurate up
152: to 2\arcsec) are given in Table~\ref{table}. Figure~\ref{fc} displays
153: a finder chart for S\,Ori\,70.
154:
155: $K_s$-band photometry and $HK$ low-resolution grism spectroscopy have
156: been collected for S\,Ori\,70 with the NIRC camera (256$\times$256
157: pixel InSb detector; Matthews \& Soifer \cite{matthews94}) at the 10-m
158: Keck I telescope on 2001 December 29--30. Meteorological conditions
159: were photometric, and the seeing was around 1\arcsec~at the time of
160: the spectroscopic observations. The $K_s$-band images consisted of
161: five co-added exposures of 15\,s each. The spectra were obtained with
162: three co-added exposures of 150\,s each using the $HK$ filter, the
163: grism gr120 and a slitwidth of 3.5\,pixel. This configuration provides
164: simultaneous coverage of the $H$- and $K$-band region from
165: 1.44\,$\mu$m to 2.5\,$\mu$m with a dispersion of
166: 5.95\,nm\,pixel$^{-1}$ and a resolution of $R$\,$\sim$\,90. Nine- and
167: three-position dither patterns were used for imaging and spectroscopy,
168: respectively. Telluric absorption was removed using the NIRC spectrum
169: of the G0 star SAO\,112561, observed at the same airmass as
170: S\,Ori\,70. Photometric calibration in the $K_s$-band in the UKIRT
171: photometric system was done with observations of L- and T-type field
172: brown dwarfs taken from Leggett et al$.$ \cite{leggett02}. We have
173: checked that our calibration of the $JHK$-bands and that of the 2MASS
174: survey are consistent within $\pm$0.05\,mag ($J$) and $\pm$0.09\,mag
175: ($H$, $K_s$). This ensures a reliable comparison to the 2MASS data of
176: ultra-cool objects (Section~\ref{membership}). Table~\ref{table}
177: provides all available photometry, and Figures~\ref{fotom}
178: and~\ref{spectrum} present color-magnitude diagrams and the final
179: smoothed $HK$-band spectrum of S\,Ori\,70.
180:
181: In this spectrum S\,Ori\,70 presents broad strong absorption features
182: that we safely identify as water vapor and methane bands. Hence, our
183: near-infrared spectroscopy confirms the ultra-cool nature of our
184: candidate: it belongs to the recently defined ``methane'' spectral
185: class T (Burgasser et al$.$ \cite{burgasser02}; Geballe et al$.$
186: \cite{geballe02}). The observed optical and near-IR colors are also
187: consistent with this classification. This is the first unambiguous
188: detection of a ``methane'' object toward the line of sight of the
189: $\sigma$\,Orionis cluster. Mart\'\i n et al$.$ \cite{martin01}
190: claimed the detection of the methane bandhead at 2.2\,$\mu$m in
191: S\,Ori\,69 (tentatively classified as T0), which is 2.8\,mag brighter
192: in the $J$-band than S\,Ori\,70. S\,Ori\,70 displays, however,
193: stronger methane bands than S\,Ori\,69, supporting its cooler
194: nature. Using water and methane molecular indices defined for the
195: spectra of field T-dwarfs (Burgasser et al$.$ \cite{burgasser02};
196: Geballe et al$.$ \cite{geballe02}), we assign a spectral type of T5.5,
197: with an uncertainty of about one subclass, to S\,Ori\,70. However, we
198: note that this classification may be unreliable (see next Section).
199:
200:
201:
202: \section{Cluster membership \label{membership}}
203:
204: Before we can attempt to determine the mass of S\,Ori\,70, we first
205: have to establish its membership in the young $\sigma$\,Orionis star
206: cluster, which is 1--8\,Myr old (Zapatero Osorio et al$.$
207: \cite{osorio02}), and at a distance of 352\,pc (Perryman et al$.$
208: \cite{perryman97}). B\'ejar et al$.$ \cite{bejar01} provide further
209: discussion on other cluster properties. Our candidate was selected
210: because it follows the photometric sequence expected for the
211: $\sigma$\,Orionis cluster, as illustrated in
212: Figure~\ref{fotom}. Additionally, its near-infrared spectrum agrees
213: with our expectations for true cluster members: it is fainter and
214: cooler than the least luminous $\sigma$\,Orionis members so far
215: identified.
216:
217: If S\,Ori\,70 did show high proper motion, this would point to it
218: being a very closeby object, and hence not a member of
219: $\sigma$\,Orionis. Therefore, we have done astrometry on our images,
220: which have a maximum epoch difference of $\sim$3\,years (see
221: Section~\ref{obs}). The position of S\,Ori\,70 was measured relative
222: to the same set of four star-like sources (Figure~\ref{fc}) around the
223: target from one epoch to the next. Our measurement accuracy is limited
224: by the fact that S\,Ori\,70 is seen close to the detection limit in
225: our first epoch $I$- and $z$-band images. We find that its proper
226: motion is too small to be detected with the available data, and can
227: thus place an upper limit of 0\arcsec.1\,yr$^{-1}$ to its
228: motion. Comparing this value with the proper motions of low-mass stars
229: in the solar neighborhood (e.g., Phan-Bao et al$.$ \cite{bao01}), we
230: conclude that S\,Ori\,70 likely lies at a distance larger than
231: $\sim$40\,pc. If it were a field T6-type dwarf, it would be at
232: 75\,$\pm$\,20\,pc; we have adopted the $JHK$ absolute magnitudes of
233: Gl\,229B (Leggett et al$.$ \cite{leggett02}; Nakajima et al$.$
234: \cite{nakajima95}), a well-known T6-type brown dwarf in the field.
235:
236: Nevertheless, T-dwarfs with $J-K_s$\,$\le$\,0.6\,mag and
237: $I-J$\,$\ge$\,3.5\,mag are rather common in the field. The all-sky
238: surveys 2MASS and SDSS are finding nearby T-class brown dwarfs at a
239: space density of 0.0042--0.05\,pc$^{-3}$ (Burgasser et al$.$
240: \cite{burgasser02}; Tsvetanov et al$.$ \cite{tsvetanov00}; Leggett et
241: al$.$ \cite{leggett00}). The recent survey of Liu et al$.$
242: \cite{liu02} is consistent with these results. However, such values
243: are likely a lower limit to the true T-dwarf space density. Due to the
244: color selection criteria and the filters used by the searches, 2MASS
245: is especially sensitive to mid-T and late-T objects, whereas SDSS can
246: detect brown dwarfs in a wider temperature range, but at shorter
247: distances. D'Antona, Oliva \& Zeppieri \cite{dantona99} found three
248: ``methane'' photometric candidates when inspecting ESO public deep
249: $IJK$ images, which covered 37\,arcmin$^2$ and were complete down to
250: $J$\,=\,23.1\,mag. One of them was spectroscopically confirmed by Cuby
251: et al$.$ \cite{cuby99}. We remark that the result of D'Antona et al$.$
252: \cite{dantona99} is expected from a substellar mass function that goes
253: as d$N$/d$M$\,$\sim$\,$M^{\alpha}$, where $\alpha$\,=\,--1. Correcting
254: for the appropriate area and limiting magnitudes, we obtain that
255: 0.08--0.2 field T-dwarfs may be contaminating our INGRID survey.
256:
257: Very recently, Chabrier \cite{chabrier02} has calculated the brown
258: dwarf IMF in the Galactic disk using state-of-the-art models for the
259: description of T-dwarfs, and normalizing the function at the bottom of
260: the main sequence. This author also included brown dwarf mass, age,
261: spatial, and binary probability distributions in his calculations, and
262: found that a mass function with a slope $\alpha$\,=\,--1 is probably
263: about the upper limit for a power-law fit of the IMF in the substellar
264: domain. Based on Chabrier's \cite{chabrier02} predictions for the
265: number of ``methane'' dwarfs from a substellar IMF described by the
266: afore mentioned power-law form, we estimate that $\sim$0.3 T-class
267: field brown dwarfs may appear in our survey. This contamination
268: (0.08--0.3 objects) is non-negligible, and might indicate that our
269: object is a field T dwarf. Nevertheless, we note that if the field
270: substellar mass function were better described by a power-law fit with
271: a lower index (e.g., Moraux, Bouvier, \& Stauffer \cite{moraux01};
272: Hillenbrand \& Carpenter \cite{hillenbrand00}) , or by a
273: lognormal-type function (see discussion in Chabrier
274: \cite{chabrier02}), the expected contamination of field ultra-cool
275: dwarfs would be smaller, making the cluster membership of S\,Ori\,70
276: more likely.
277:
278: In addition to the analysis of the contamination by field T dwarfs in
279: our survey, we have compared our observed spectra of S\,Ori\,70 and of
280: the T6V field standard brown dwarf 2MASSI\,J0243137--245329 (Burgasser
281: et al$.$ \cite{burgasser02}) to a set of synthetic spectral energy
282: distributions, which cover a wide range of temperatures
283: (500\,$\le$\,$T_{\rm eff}$\,$\le$\,1800\,K) and gravities
284: (3.0$\le$\,\,log\,$g$\,$\le$\,5.0, in CGS units). These theoretical
285: spectra, kindly provided by F$.$ Allard, are described in Allard et
286: al$.$ \cite{allard01}. Basically, we have adopted the {\sc ``cond''}
287: models, which treat dust as if condensed in deep layers of the
288: photosphere. It is believed that dust is no longer working as an
289: opacity source at near-infrared wavelengths in T-class atmospheres
290: (Tsuji, Ohnaka, \& Aoki \cite{tsuji99}; Allard et al$.$
291: \cite{allard01}; Burrows, Marley, \& Sharp \cite{burrows00}).
292: Observed and synthetic spectra were compared through a least square
293: minimization technique to determine the best-fitting $T_{\rm eff}$ and
294: log\,$g$ parameters for our two objects. This is done by minimizing
295: the sum ($\sigma^{2}$) of the squares of the differences between the
296: flux values of the program objects and the theoretical models. We
297: looked for least squares matches based on the data sets in the
298: intervals of 1.50--1.69\,$\micron$ and 1.99--2.40\,$\micron$, which
299: include molecular bands from methane, water vapor, and carbon
300: monoxide. Other wavelengths, with poorer signal-to-noise in their
301: fluxes and possibly affected by telluric absorption, were not
302: considered. Figure~\ref{logg} depicts $\sigma^{2}$ as a function of
303: the range of $T_{\rm eff}$ and log\,$g$ in the theoretical spectra.
304: We find that the best fit to the T6V 2MASS field standard brown dwarf
305: yields $T_{\rm eff}$\,=\,950\,$\pm$\,50\,K and
306: log\,$g$\,=\,5.0\,$\pm$\,0.5 (cm\,s$^{-2}$), in perfect agreement with
307: Gl\,229B (Marley et al$.$ \cite{marley96}; Allard et al$.$
308: \cite{allard96}; Oppenheimer et al$.$ \cite{oppenheimer98}).
309:
310: The $HK$ spectrum of S\,Ori\,70, on the other hand, is better
311: reproduced by cooler and lower gravity models according to the least
312: square minimization technique. We find the best fits for $T_{\rm
313: eff}$\,=\,800\,$^{+200}_{-100}$\,K and log\,$g$\,=\,4.0\,$\pm$\,1.0
314: (cm\,s$^{-2}$). The observed spectra of S\,Ori\,70 and
315: 2MASSI\,J0243137--245329, along with a few theoretical spectra are
316: shown in Figure~\ref{models}. The lower gravity of S\,Ori\,70 compared
317: to the field T6V dwarf again supports its youth. Furthermore,
318: evolutionary models predict that, at the age of the $\sigma$\,Orionis
319: cluster, objects with similar temperatures and colors have logarithmic
320: gravities around 3.0--4.0, consistent with our determination.
321:
322: The colors of S\,Ori\,70 fall within the range of measurements for T
323: dwarfs, as shown in Figure~16 of Burgasser et al$.$
324: \cite{burgasser02}. However, this object presents optical and
325: near-infrared colors that are somewhat different from those of field
326: dwarfs with late-T spectral types. We exclude the $z$-band from these
327: comparisons due to the lack of an absolute calibration for our
328: data. Table~\ref{color} lists differences with respect to the spectral
329: classes T6V, T7V, and T8V, which have $T_{\rm eff}$ similar to that of
330: S\,Ori\,70. Averaged 2MASS $JHK$ colors for the field have been taken
331: from Burgasser et al$.$ \cite{burgasser02}, and the $I$ color belongs
332: to Gl\,229B (T6V; Matthews et al$.$ \cite{matthews96}). Color
333: transformations between UKIRT and 2MASS photometric systems (Carpenter
334: \cite{carpenter01}) are significantly smaller than the relatively
335: large uncertainties in the photometry of S\,Ori\,70. We can explain
336: these differences in terms of low gravity. According to theoretical
337: computations, water absorptions in the near-infrared are stronger for
338: lower surface gravities, while methane absorptions appear weaker
339: (Allard et al$.$ \cite{allard96}). At optical wavelengths, the
340: effects of gravity are noticeable in the strength of atomic lines
341: (essentially K\,{\sc i} and Na\,{\sc i}). The second column of
342: Table~\ref{color} provides color differences for a model of 900\,K;
343: the high gravity model (log\,$g$\,=\,5.0) has been subtracted from the
344: low gravity one (log\,$g$\,=\,3.0). Such dissimilarities are
345: comparable to those of S\,Ori\,70, supporting the object's low gravity
346: atmosphere.
347:
348: On the basis of all the available photometric and spectroscopic
349: properties of S\,Ori\,70, in addition to the relatively low
350: contamination expected for T-class field interlopers ($\le$\,0.3
351: objects), we conclude that S\,Ori\,70 is a possible member of the
352: $\sigma$\,Orionis cluster.
353:
354:
355: \section{Mass determination and final remarks \label{mass}}
356:
357: In the following analysis, we derive the properties of S\,Ori\,70
358: assuming that it is a member of the $\sigma$\,Orionis cluster. Its
359: mass is obtained through comparison with state-of-the-art evolutionary
360: models. This exercise is complicated by the marked dependence of
361: T-class colors on gravity (see Table~\ref{color}). Overplotted onto
362: the data of Figure~\ref{fotom} are two isochrones for 1 and 10\,Myr
363: from Baraffe et al$.$ \cite{baraffe02}. Theoretical effective
364: temperatures and luminosities have been transformed into observables
365: using temperature calibrations and bolometric corrections (Leggett et
366: al$.$ \cite{leggett02}; Basri et al$.$ \cite{basri00al}; Reid et al$.$
367: \cite{reid01a}), which have been calculated for field, high-gravity
368: dwarfs. As shown in Figure~\ref{fotom}, this is a reasonable approach
369: for warmer spectral types since colors and bolometric corrections are
370: similar for different gravities. Allard et al$.$ \cite{allard01},
371: \cite{allard96} noted that for low surface temperatures (below
372: 1500\,K, i.e., T-types), optical wavelengths and the $K$-band are
373: strongly affected by the gravity change. This could explain the
374: apparent lack of consistency between the tracks and the location of
375: S\,Ori\,70 in Figure~\ref{fotom}. On the other hand, S\,Ori\,70 may be
376: a binary, thus appearing overluminous in our Figure. Such an
377: interpretation cannot be excluded, since Mart\'\i n et al$.$
378: \cite{martin00} argued that the photometric binary fraction of
379: Pleiades brown dwarfs is about 25\%, which is similar to the
380: percentage of binaries found among field brown dwarfs (Reid et al$.$
381: \cite{reid01b}).
382:
383: From Table~\ref{color} we infer that the bolometric correction in the
384: $J$-band is not so much affected by gravity. This allows us to derive
385: the luminosity of S\,Ori\,70 to a certain degree of confidence. We
386: find it to be log\,$L/L_{\odot}$\,=\,--4.13\,$\pm$\,0.15. For this,
387: we have used the averaged $J$ bolometric correction of T6V and T8V
388: field brown dwarfs from Leggett et al$.$
389: \cite{leggett02}. Figure~\ref{edad} displays the luminosity evolution
390: of objects of different masses according to the models by Chabrier et
391: al$.$ \cite{chabrier00} and Burrows et al$.$
392: \cite{burrows97}. S\,Ori\,70 is indicated with a thick line that takes
393: into account the age interval adopted for the $\sigma$\,Orionis
394: cluster (1--8\,Myr). The mass of our ``methane'' object can be derived
395: from the Figure at 3\,$^{+5}_{-1}$\Mj, where we have adopted 3\,Myr as
396: the most likely cluster age (Wolk \cite{wolk96}; B\'ejar, Zapatero
397: Osorio, \& Rebolo \cite{bejar99}; Zapatero Osorio et al$.$
398: \cite{osorio02}; Oliveira et al$.$ \cite{oliveira02}). The upper and
399: lower error bars take into account differences between models, the age
400: range of the cluster, and the uncertainty in luminosity. Nevertheless,
401: we shall caution that the reliability of this mass estimate relies on
402: the evolutionary tracks, and thus is subject to possible systematic
403: effects arising from the description of atmospheres. Our derived mass
404: is compatible within uncertainties with the estimated $T_{\rm eff}$
405: and log\,$g$. In relative terms, if S\,Ori\,70 is a member of the
406: $\sigma$\,Orionis cluster, it is the least massive object so far
407: identified in isolation beyond our solar system.
408:
409: Finding S\,Ori\,70 in the small area covered by our near-infrared
410: survey might have some impact on our knowledge of the
411: $\sigma$\,Orionis substellar IMF, despite the dubious reliability
412: based on a single object. Far from suggesting a decrease in the number
413: of objects with planetary masses below 5--8\Mj~(Lucas \& Roche
414: \cite{lucas00}; Muench et al$.$ \cite{muench02}), the discovery of
415: S\,Ori\,70 is consistent with an extension of the IMF toward very low
416: masses that may still continue with a slope of $\alpha$\,$\sim$\,--1
417: (d$N$/d$M$\,$\sim$\,$M^{\alpha}$) in the planetary mass domain. This
418: represents the natural extension of the brown dwarf IMF given in
419: B\'ejar et al$.$ \cite{bejar01} for the $\sigma$\,Orionis cluster.
420: However, we remark that our survey only imaged a very small fraction
421: (0.2\%) of the total cluster area, and that {\it one} object has very
422: low statistical meaning. Hence, larger area searches need to be
423: conducted for a better knowledge of the $\sigma$\,Orionis planetary
424: IMF.
425:
426: The detection of S\,Ori\,70 indicates that objects only slightly
427: heavier than Jupiter may exist free-floating in
428: $\sigma$\,Orionis. Their formation process is not yet
429: established. Theory predicts that opacity-limited fragmentation of
430: cool gravitationally collapsing gas clouds is capable of producing
431: 7--10\Mj~Population~I objects in isolation (Silk \cite{silk77}; Low \&
432: Lynden-Bell \cite{low76}). Moreover, this minimum Jeans mass seems to
433: be insensitive to changes in the opacity of protostellar clouds
434: (amount of dust, size of grains, cosmic ray flux; Boss
435: \cite{boss88}). These models, however, do not include rotation,
436: magnetic fields, and further external accretion onto the cloud
437: fragment, which might alter the final mass of the nascent
438: object. S\,Ori\,70 is probably less massive than the minimum Jeans
439: mass of 7--10\Mj, and thus prompts us to refine the
440: collapse-and-fragmentation models and/or to rethink possible formation
441: mechanisms for such low-mass objects. Recently, several formation
442: scenarios have been suggested that include tidal interactions and
443: ejection of low-mass objects from multiple systems before brown dwarfs
444: and planetary-mass objects can accrete enough gas to become stars
445: (e.g., Reipurth \& Clarke \cite{reipurth01}; Boss \cite{boss01}; Bate,
446: Bonnell, \& Bromm \cite{bate02}). S\,Ori\,70 is located 8\arcmin.65 to
447: the west of the multiple stellar system $\sigma$\,Orionis, which is
448: formed by six hot massive stars (Morrell \& Levaton \cite{morrell91};
449: Bolton \cite{bolton74}). S\,Ori\,70 might have been ejected from it
450: with a space velocity of
451: $v$\,sin\,$\gamma$\,$\sim$\,0.29\,km\,s$^{-1}$ for a cluster age of
452: 3\,Myr, where $\gamma$ stands for the ejection direction with respect
453: to the line of sight. On the other hand, Padoan \& Nordlund
454: \cite{padoan02}, and references therein, suggest that brown dwarfs are
455: formed in the same way as more massive hydrogen burning stars, that is
456: by the process of supersonic turbulent fragmentation. These authors
457: claim that their model for the IMF predicts a number of substellar
458: objects consistent with observational data. If any or all of these
459: scenarios are viable solutions to the formation problem can only be
460: decided when more such low-mass objects are known, and their
461: properties have been studied in detail.
462:
463:
464: \acknowledgments We are grateful to F$.$ Allard and I$.$ Baraffe for
465: providing us computer-ready files of their models. We are also
466: grateful to the anonymous referee for valuable comments. This paper is
467: based on observations obtained at the WHT, operated by the Isaac
468: Newton Group of Telescopes, funded by PPARC at the Spanish
469: Observatorio del Roque de los Muchachos of the Instituto de Astrof\'\i
470: sica de Canarias; at the W.\,M$.$ Keck Observatory, which is operated
471: as a scientific partnership among the California Institute of
472: Technology, the University of California and the National Aeronautics
473: and Space Administration (the Observatory was made possible by the
474: generous financial support of the W.\,M. Keck Foundation); and at the
475: Carlos S\'anchez Telescope, at the Spanish Teide Observatory of the
476: Instituto de Astrof\'\i sica de Canarias. This publication makes use
477: of data products from the Two Micron All Sky Survey, which is a joint
478: project of the University of Massachusetts and the Infrared Processing
479: and Analysis Center/California Institute of Technology, funded by the
480: National Aeronautics and Space Administration and the National Science
481: Foundation. Partial financial support comes from the Spanish
482: ``Programa Ram\'on y Cajal''.
483:
484:
485:
486: \begin{thebibliography}{}
487: \bibitem[2001]{allard01} Allard, F., Hauschildt, P.\,H., Alexander,
488: D.\,R., Tamanai, A., \& Schweitzer, A$.$ 2001, \apj, 556, 357
489: \bibitem[1996]{allard96} Allard, F., Hauschildt, P.\,H., Baraffe,
490: I. \& Chabrier, G$.$ 1996, \apj, 465, L123
491: \bibitem[2002]{baraffe02} Baraffe, I., Chabrier, G., Allard, F.,
492: \& Hauschildt, P.\,H$.$ 2002, \aap, 382, 563
493: \bibitem[2000]{basri00} Basri, G$.$ 2000, \araa, 38, 485
494: \bibitem[2000]{basri00al} Basri, G., Mohanty, S., Allard, F.,
495: Hauschildt, P.\,H., Delfosse, X., Mart\'\i n, E.\,L., Forveille, T.,
496: \& Goldman, B$.$ 2000, \apj, 538, 363
497: \bibitem[2002]{bate02} Bate, M.\,R., Bonnell, I.\,A.,\& Bromm, V$.$
498: 2002, \mnras, in press
499: \bibitem[2002]{bejar02} B\'ejar, V.\,J.\,S., et al$.$ 2002, in
500: preparation
501: \bibitem[2001]{bejar01} B\'ejar, V.\,J.\,S., et al$.$ 2001, \apj,
502: 556, 830
503: \bibitem[1999]{bejar99} B\'ejar, V.\,J.\,S., Zapatero Osorio, M.\,R.,
504: \& Rebolo, R$.$ 1999, ApJ, 521, 671
505: \bibitem[1974]{bolton74} Bolton, C.\,T$.$ 1974, \apj, 192, L7
506: \bibitem[1988]{boss88} Boss, A.\,P. 1988, \apj, 331, 370
507: \bibitem[2001]{boss01} Boss, A.\,P. 2001, \apj, 551, L167
508: \bibitem[2002]{burgasser02} Burgasser, A.\,J., et al$.$ 2002, \apj,
509: 564, 421
510: \bibitem[1997]{burrows97} Burrows, A., Marley, M., Hubbard, W.\,B.,
511: Lunine, J.\,I., Guillot, T., Saumon, D., Freedman, R., Sudarsky, D.,
512: \& Sharp, C$.$ 1997, \apj, 491, 856
513: \bibitem[2000]{burrows00} Burrows, A., Marley, M.\,S., \& Sharp,
514: C.\,M$.$ 2000, \apj, 531, 438
515: \bibitem[2001]{carpenter01} Carpenter, J.\,M$.$ 2001, \aj, 121, 2851
516: \bibitem[2002]{chabrier02} Chabrier, G$.$ 2002, \apj, 567, 304
517: \bibitem[2000]{chabrier00} Chabrier, G., Baraffe, I., Allard, F., \&
518: Hauschildt, P.\,H$.$ 2000, \apj, 542, 464
519: \bibitem[1999]{cuby99} Cuby, J.\,G., et al$.$ 1999, \aap, 349, L41
520: \bibitem[1999]{dantona99} D'Antona, F., Oliva, E., \& Zeppieri, A$.$
521: 1999, \aap, 352, 567
522: \bibitem[2002]{geballe02} Geballe, T.\,R., et al$.$ 2002, \apj,
523: 564, 466
524: \bibitem[2000]{hillenbrand00} Hillenbrand, L.\,A., \& Carpenter,
525: J.\,M$.$ 2000, \apj, 540, 236
526: %\bibitem[1999]{hambly99} Hambly, N.\,C., Hodgkin, S.\,T.,
527: Cossburn, M.\,R., \& Jameson, R.\,F$.$ 1999, \mnras, 303, 835
528: \bibitem[1998]{hunt98} Hunt, L.\,K., Mannucci, F., Testi, L.,
529: Migliorini, S., Stanga, R.\,M., Baffa, C., Lisi, F., \& Vanzi,
530: L$.$ 1998, \aj, 115, 2594
531: \bibitem[1963]{kumar63} Kumar, S.\,S. 1963, \apj, 137, 1121
532: %\bibitem[1986]{larson86} Larson, R.\,B$.$ 1986, \mnras, 218, 409
533: \bibitem[2000]{leggett00} Leggett, S.\,K., et al$.$ 2000, \apj, 536,
534: L35
535: \bibitem[2002]{leggett02} Leggett, S.\,K., et al$.$ 2002, \apj, 564,
536: 452
537: \bibitem[2002]{liu02} Liu, M.\,C., Wainscoat, R., Mart\'\i n, E.\,L.,
538: Barris, B., \& Tonry, J$.$ 2002, \apj, in press (astro-ph/0202508)
539: \bibitem[1976]{low76} Low, C., \& Lynden-Bel, D$.$ 1976, \mnras, 176,
540: 367
541: \bibitem[2000]{lucas00} Lucas, P.\,W., \& Roche, P.\,F$.$ 2000, \mnras,
542: 314, 858
543: \bibitem[2001]{lucas01} Lucas, P.\,W., Roche, P.\,F., Allard, F., \&
544: Hauschildt, P.\,H$.$ 2001, \mnras, 326, 695
545: \bibitem[2000]{luhman00} Luhman, K.\,L., Rieke, G.\,H., Young,
546: E.\,T., Cotera, A.\,S., Chen, H., Rieke, M.\,J., Schneider, G.,
547: \& Thompson, R.\,I$.$ 2000, \apj, 540, 1016
548: \bibitem[1996]{marley96} Marley, M.\,S., Saumon, D., Guillot, T.,
549: Freedman, R.\,S., Hubbard, W.\,B., Burrows, A., \& Lunine, J.\,I$.$
550: 1996, Science, 272, 1919
551: \bibitem[2000]{martin00} Mart\'\i n, E.\,L., Brandner, W., Bouvier, J.,
552: Luhman, K.\,L., Stauffer, J., Basri, G., Zapatero Osorio, M.\,R., \&
553: Barrado y Navascu\'es, D$.$ 2000, \apj, 543, 299
554: \bibitem[2001]{martin01} Mart\'\i n, E.\,L., Zapatero Osorio, M.\,R.,
555: Barrado y Navascu\'es, D., B\'ejar, V.\,J.\,S., \& Rebolo, R$.$
556: 2001, \apj, 558, L117
557: \bibitem[1996]{matthews96} Matthews, K., Nakajima, T., Kulkarni,
558: S.\,K., \& Oppenheimer, B.\,R$.$ 1996, \aj, 112m 1678
559: \bibitem[1994]{matthews94} Matthews, K., \& Soifer, B.\,T$.$ 1994,
560: Exp$.$ Astron., 3, 77
561: \bibitem[2001]{moraux01} Moraux, E., Bouvier, J., \& Stauffer,
562: J.\,R$.$ 2001, \aap, 367, 211
563: \bibitem[1991]{morrell91} Morrell, N, \& Levato, H$.$ 1991, \apjs,
564: 75, 965
565: \bibitem[2002]{muench02} Muench, A., Lada, E., Lada, C., \&
566: Alv\'es, J$.$ 2002, \apj, in press
567: \bibitem[1995]{nakajima95} Nakajima, T., Oppenheimer, B.\,R.,
568: Kulkarni, S.\,R., Golimowski, D.\,A., Matthews, K., \& Durrance,
569: S.\,T$.$ 1995, \nat, 378, 463
570: \bibitem[1995]{oke95} Oke, J.\,B., et al$.$ 1995, \pasp, 107, 375
571: \bibitem[2002]{oliveira02} Oliveira, J.\,M., Jeffries, R.\,D., Kenyon,
572: M.\,J., Thompson, S.\,A., \& Naylor, T$.$ 2002, A\&A, 382, L22
573: \bibitem[1998]{oppenheimer98} Oppenheimer, B.\,R., Kulkarni, S.\,R.,
574: Matthews, K., \& Nakajima, T$.$ 1995, Science, 270, 1478
575: \bibitem[2002]{padoan02} Padoan, P., \& Nordlund, \aa$.$ 2002, \apj,
576: submitted
577: \bibitem[2001]{bao01} Phan-Bao, N., Guibert, J., Crifo, F.,
578: Delfosse, X., Forveille, T., Borsenberger, J., Epchtein, N.,
579: Fouqu\'e, P., \& Simon, G$.$ 2001, \aap, 380, 590
580: \bibitem[1997]{perryman97} Perryman, M.\,A.\,C., et al$.$ 1997, A\&A,
581: 323, L49
582: \bibitem[2001a]{reid01a} Reid, I.\,N., Burgasser, A.\,J., Cruz, K.\,L.,
583: Kirkpatrick, J.\,D., \& Gizis, J.\,E$.$ 2001a, \aj, 121, 1710
584: \bibitem[2001b]{reid01b} Reid, I.\,N., Gizis, J.\,E.; Kirkpatrick,
585: J.\,D., \& Koerner, D.\,W$.$ 2001b, \aj, 121, 489
586: \bibitem[2001]{reipurth01} Reipurth, B., \& Clarke, C. 2001, \aj, 122,
587: 432
588: \bibitem[1996]{saumon96} Saumon, D., Hubbard, W.\,B., Burrows, A.,
589: Guillot, T., Lunine, J.\,I., \& Chabrier, G$.$ 1996, \apj, 460, 993
590: \bibitem[1977]{silk77} Silk, J$.$ 1977, \apj, 214, 152
591: \bibitem[1999]{tsuji99} Tsuji, T., Ohnaka, K., \& Aoki, W$.$ 1999,
592: \apj, 520, L119
593: \bibitem[2000]{tsvetanov00} Tsvetanov, Z.\,I., et al$.$ 2000, \apj,
594: 531, L61
595: \bibitem[1996]{wolk96} Wolk, S.\,J$.$ 1996, Ph.\,D$.$ thesis, Univ$.$
596: New York at Stony Brook
597: \bibitem[1999a]{osorio99a} Zapatero Osorio, M.\,R., B\'ejar,
598: V.\,J.\,S., Rebolo, R., Mart\'\i n, E.\,L., \& Basri, G$.$ 1999a,
599: \apj, 524, L115
600: \bibitem[2000]{osorio00} Zapatero Osorio, M.\,R., B\'ejar, V.\,J.\,S.,
601: Mart\'\i n, E.\,L., Rebolo, R., Barrado y Navascu\'es, D.,
602: Bailer-Jones, C.\,A.\,L., \& Mundt, R$.$ 2000, Science, 290, 103
603: \bibitem[2002]{osorio02} Zapatero Osorio, M.\,R., B\'ejar, V.\,J.\,S.,
604: Pavlenko, Ya., Rebolo, R., Allende Prieto, C., Mart\'\i n, E.\,L., \&
605: Garc\'\i a L\'opez, R.\,J$.$ 2002, \aap, 384, 937
606: \bibitem[1999b]{osorio99b} Zapatero Osorio, M.\,R., Rebolo, R.,
607: Mart\'\i n, E.\,L., Hodgkin, S.\,T., Cossburn, M.\,R., Magazz\`u, A.,
608: Steele, I.\,A., \& Jameson, R.\,R$.$ 1999b, \aaps, 134, 537
609:
610: \end{thebibliography}
611:
612: \clearpage
613: \begin{figure}
614: \plotone{f1.eps}
615: \caption{S\,Ori\,70 finder chart (NIRC $K_s$ image,
616: 32\arcsec$\times$32\arcsec). \label{fc}}
617: \end{figure}
618:
619: \clearpage
620: \begin{figure}
621: \plottwo{f2.eps}{f3.eps}
622: \caption{S\,Ori\,70 extrapolates the optical and near-infrared
623: photometric sequence of the $\sigma$\,Orionis cluster; other cluster
624: members (B\'ejar et al$.$ \cite{bejar99}, \cite{bejar01}; Zapatero
625: Osorio et al$.$ \cite{osorio02}) are indicated with open
626: circles. Theoretical evolutionary models for 1\,Myr (solid line) and
627: 10\,Myr (dotted line) are overplotted onto the data (Baraffe et al$.$
628: \cite{baraffe02}), and masses in units of Jupiter mass are also
629: provided. \label{fotom}}
630: \end{figure}
631:
632: \clearpage
633: \begin{figure}
634: \plotone{f4.eps}
635: \caption{NIRC spectrum of S\,Ori\,70 (smoothed with a boxcar of 5
636: pixel). The two spectral standard field brown dwarfs
637: SDSS\,J0857585+570851, with spectral type L8V, and
638: 2MASSI\,J0243137--245329, with spectral type T6V, obtained with NIRC
639: on the same night as S\,Ori\,70, are shown for comparison. All spectra
640: are normalized to their peak flux in the $H$ band, and have been
641: shifted by 1 unit for clarity. Most relevant features are
642: indicated. \label{spectrum}}
643: \end{figure}
644:
645: \clearpage
646: \begin{figure}
647: \plotone{f5.eps}
648: \caption{Residuals of fits of the observed near-infrared $HK$ spectrum
649: of S\,Ori\,70 (upper panel) and of the T6V-type field brown dwarf
650: 2MASSI\,J0243137--245329 (lower panel) to synthetic spectra covering a
651: range of $T_{\rm eff}$ and log\,$g$. The best-fitting models are
652: marked by arrows for both objects. The synthetic spectra assume solar
653: composition (Allard et al$.$ \cite{allard01}). Effective temperatures
654: (K) are indicated. Logarithmic gravities are in
655: cm\,s$^{-2}$.\label{logg}}
656: \end{figure}
657:
658: \clearpage
659: \begin{figure}
660: \plottwo{f6.eps}{f7.eps}
661: \caption{Comparison of observed spectra to theoretical {\sc ``cond''}
662: spectra (Allard et al$.$ \cite{allard01}), which have been degraded to
663: match the resolution of the observations. Logarithmic gravities are in
664: cm\,s$^{-2}$. Spectral features and normalization are as in
665: Figure~\ref{spectrum}.\label{models}}
666: \end{figure}
667:
668: \clearpage
669: \begin{figure}
670: \plottwo{f8.eps}{f9.eps}
671: \caption{Luminosity evolution of very low-mass objects according to
672: theoretical models by Chabrier et al$.$ \cite{chabrier00} and Burrows
673: et al$.$ \cite{burrows97}. The location of S\,Ori\,70, considering all
674: possible cluster ages, is given by the thick line. The most likely
675: cluster age is estimated at 3\,Myr (vertical dashed line). The
676: short-dashed lines indicate the error in luminosity. Masses are units
677: of Jupiter mass.\label{edad}}
678: \end{figure}
679:
680: \clearpage
681: \begin{deluxetable}{cccccccccccc}
682: \tabletypesize{\scriptsize}
683: \tablecaption{Data for S\,Ori\,70. \label{table}}
684: \tablewidth{0pt}
685: \tablehead{
686: \colhead{ } &
687: \colhead{RA} &
688: \colhead{DEC} &
689: \colhead{ } &
690: \colhead{ } &
691: \colhead{ } &
692: \colhead{ } &
693: \colhead{ } &
694: \colhead{$T_{\rm eff}$} &
695: \colhead{log\,$g$} &
696: \colhead{ } \\
697: \colhead{IAU name} &
698: \colhead{(J2000)} &
699: \colhead{(J2000)} &
700: \colhead{$J$} &
701: \colhead{$J-H$} &
702: \colhead{$J-K_s$} &
703: \colhead{$I-J$} &
704: \colhead{$I-z$} &
705: \colhead{(K)} &
706: \colhead{(cm\,s$^{-2}$)} &
707: \colhead{SpT}
708: }
709: \startdata
710: S\,Ori\,J053810.1--023626 & 05 38 10.1 & --02 36 26 &
711: 20.28$\pm$0.10 & --0.14$\pm$0.15 & 0.50$\pm$0.20 &
712: 4.75$\pm$0.40 & 2.2$\pm$0.4 & 700--1000 & 4.0$\pm$1.0 & T class\\
713: \enddata
714: \tablecomments{Units of right ascension are hours, minutes, and
715: seconds, and units of declination are degrees, arcminutes, and
716: arcseconds. Coordinates are accurate to $\pm$2\arcsec. See text
717: for the calibration of the $z$-band.}
718: \end{deluxetable}
719:
720: \begin{deluxetable}{cccccc}
721: \tabletypesize{\scriptsize}
722: \tablecaption{Color differences \label{color}}
723: \tablewidth{0pt}
724: \tablehead{
725: \colhead{} &
726: \colhead{ } &
727: \colhead{S\,Ori\,70} &
728: \colhead{S\,Ori\,70} &
729: \colhead{S\,Ori\,70} \\
730: \colhead{} &
731: \colhead{Model\tablenotemark{a}} &
732: \colhead{-- T6V} &
733: \colhead{-- T7V} &
734: \colhead{-- T8V}
735: }
736: \startdata
737: $I-J$ & $-$0.47 & $-$0.9 & \nodata & \nodata \\
738: $J-H$ & $-$0.10 & $-$0.21 & $+$0.04 & $-$0.23 \\
739: $H-K$ & $+$0.49 & $+$0.83 & $+$0.31 & $+$0.59 \\
740: $J-K$ & $+$0.38 & $+$0.53 & $+$0.35 & $+$0.36 \\
741: \enddata
742: \tablenotetext{a}{{\sc ``cond''} 900\,K,
743: [log\,$g$\,=\,3.0]\,--\,[log\,$g$\,=\,5.0]}
744: \end{deluxetable}
745:
746:
747: \end{document}
748:
749: