1: \documentclass[12pt,preprint]{aastex}
2: %\documentclass[apj]{emulateapj}
3: %\usepackage{apjfonts}
4:
5: \shorttitle{Wide Substellar Companion to an M-dwarf}
6: \shortauthors{Radigan et al.}
7:
8: \newcommand{\msun}{\ensuremath{M_{\odot}}}
9: \newcommand{\mjup}{\ensuremath{M_{\rm Jup}}}
10:
11: \begin{document}
12:
13: \title{Discovery of a Wide Substellar Companion to a Nearby Low-Mass
14: Star}
15:
16: \author{Jacqueline Radigan, David Lafreni\`ere, Ray Jayawardhana}
17: \affil{Department of Astronomy and Astrophysics, University
18: of Toronto, 50 St. George Street, Toronto, ON, M5S 3H4, Canada}
19: \author{Ren\'e Doyon}
20: \affil{D\'epartement de Physique and Observatoire du Mont M\'egantic, Universit\'e de Montr\'eal, C.P. 6128, Succ. Centre-Ville, Montr\'eal, QC, H3C 3J7, Canada}
21: \email{radigan@astro.utoronto.ca}
22:
23: \begin{abstract}
24:
25: We report the discovery of a wide ($135\pm25$~AU), unusually blue L5
26: companion 2MASS J17114559+4028578 to the
27: nearby M4.5 dwarf G 203-50 as a result of a targeted search for common
28: proper motion pairs in the Sloan Digital Sky Survey and the Two
29: Micron All Sky Survey. Adaptive Optics imaging with Subaru indicates that neither
30: component is a nearly equal mass binary with separation $> 0.18\arcsec$, and
31: places limits on the existence of additional faint companions. An examination of TiO and CaH features
32: in the primary's spectrum is consistent with solar metallicity and
33: provides no evidence that G 203-50 is metal poor. We estimate an
34: age for the primary of 1-5 Gyr based on activity. Assuming
35: coevality of the companion, its age, gravity and metallicity can be
36: constrained from properties of the primary, making it a suitable
37: benchmark object for the calibration of evolutionary models and for
38: determining the atmospheric properties of peculiar blue L dwarfs.
39: The low total mass ($M_{tot}=0.21\pm0.03$~\msun), intermediate mass
40: ratio ($q=0.45\pm0.14$), and wide separation of this system demonstrate that the star
41: formation process is capable of forming wide, weakly bound binary systems
42: with low mass and BD components. Based on the sensitivity
43: of our search we find that no more than $2.2\%$ of early-to-mid M dwarfs ($9.0
44: <M_V < 13.0$) have wide substellar companions with $m>0.06$~\msun.
45:
46: \end{abstract}
47:
48: \keywords{binaries: general --- stars: formation --- stars: individual
49: (2MASS J17114559+4028578, G 203-50) stars: low-mass, brown dwarfs}
50:
51: \section{Introduction}\label{sect:intro}
52:
53: Although star birth is a complex process, the
54: observation of binary systems--frequencies, mass ratios,
55: and separations--can provide insight into the formation process as well as
56: constraints for theoretical models.
57: The formation of brown dwarfs (BDs) is particularly challenging since their masses are an order of
58: magnitude smaller than the typical Jeans mass in molecular clouds.
59:
60: Whether BDs form similarly to their more massive stellar counterparts
61: or require additional mechanisms is currently an open question. The answer may lie in the multiplicity
62: properties of these substellar objects.
63: Whereas free-floating BDs are observed in abundance, finding BDs as companions to stars has
64: proved more difficult. A ``brown dwarf desert'' ($\lesssim0.5\%$ companion fraction) is observed at
65: close separations ($<3$~AU) to main sequence stars, in comparison to
66: a significant number of both planetary and stellar mass companions
67: seen at similar separations \citep{marcy00}. It has recently been
68: determined that this desert does not extend out to larger separations for solar analogs (F,G,K stars),
69: $\sim7\%$ of which are found to harbor substellar companions at
70: separations greater than 30~AU \citep{met05}.
71: However, searches for substellar companions to M dwarfs at large
72: separations ($\gtrsim$ 40 AU) have yielded mostly null results
73: \citep[e.g.][]{allen08,mccarthy04,hinz02,daemgen07} or sparse results
74: \citep[e.g.][]{oppen01}. There are only 5 known BD
75: companions to stellar M dwarf primaries at separations greater than 40AU: TWA 5b,c
76: \citep{lowrance99}, G 196-3B \citep{rebolo98}, GJ 1001B
77: \citep{goldman99}, Gl 229B \citep{nakajima95}, and LP 261-75B
78: \citep{burgasser05,reid06}. For thoroughness we note that the L2.5 companion GJ 618.1B
79: \citep{wilson01} may also fall into this category, however it is
80: more likely stellar. In the VLM regime
81: ($M_{1}<0.1$~\msun) surveys have found that no more than $\sim1\%$ of
82: stars have wide companions, including stellar ones
83: \citep{burgasser07a}. Additionally, VLM binaries are found to be on
84: average 10-20 times more tightly bound than their stellar counterparts, hinting
85: that disruptive dynamical interactions may play an important role in
86: their formation \citep{close03}. These observations have been cited as evidence in favor of
87: the ejection hypothesis \citep{rep01, bate05} where BDs and VLM stars
88: are thought to be stellar embryos formed by the fragmentation of a
89: more massive pre-stellar core, then prematurely ejected from their
90: birth environments. However, BD companions to more massive stars do not
91: tend to form harder binaries than stellar systems of similar total
92: mass \citep[e.g.][]{reid01,met05}. While this is potentially evidence
93: that BDs can form similarly to stars via turbulent fragmentation
94: within molecular clouds \citep{padoan02}, it is also consistent with simulations of
95: disk instabilities \citep[e.g.][]{stamatellos07,boss00}, which are capable of producing
96: substellar companions around more massive primaries.
97:
98: While a significant fraction of solar mass stars may retain wide
99: BD companions, this does not seem to hold true for lower mass stars.
100: As a result, very few wide BD companions to low mass stars are known.
101: The discovery and characterization of these systems, especially in the intermediate
102: range between the solar analog and VLM regimes, will help complete the
103: emerging picture of BD multiplicity at wide separations.
104: Additionally, wide BD companions to stars make suitable
105: ``benchmark'' objects, as their properties can be
106: inferred from those of the primary \citep{pinfield06}. This is important for the
107: calibration of BD evolutionary models, which requires independent age estimates.
108:
109: Here we present the discovery of a wide substellar companion to a nearby M4.5
110: star. The search, discovery and followup observations are outlined in \S\ref{sect:search}, while the
111: physical properties of the system and its components are given in \S
112: \ref{sect:properties}. In \S\ref{sect:discussion} we
113: discuss the companion's unusual NIR colors, possible formation
114: scenarios, and the sensitivity of our search. Given a space
115: density for M dwarfs we make a crude estimate of how rare such systems
116: may be. A brief summary and outlook are presented in \S \ref{sect:conclusions}.
117:
118:
119: \section{Discovery and Observations}\label{sect:search}
120:
121: The binary G~203-50 / 2MASS J17114559+4028578 (G~203-50AB hereafter)
122: was found in a cross-match of the SDSS DR6 Photoprimary Catalogue
123: \citep{sdss} and
124: 2MASS Point Source Catalogue \citep{skrutskie06} in which we searched for common proper
125: motion pairs containing at least one VLM or BD component. The
126: cross-correlation of catalogues, calculation of proper motions, and the
127: identification of co-moving stars was done in parallel for 4~$\rm{deg^2}$
128: sections of the sky at a time, spanning the contiguous region of the
129: SDSS Legacy survey in the northern galactic cap. We made preliminary cuts to include only 2MASS
130: sources with S/N $>$ 5 in at least one band (J, H, or K) and not
131: flagged as minor planets, and SDSS
132: sources that were not classified as ``sky pointings'' or electronic
133: ghosts. For every 2MASS source the closest SDSS match was found and
134: proper motion vectors with uncertainties were computed. A cut was made in order to select only
135: stars that had moved at the $3\sigma$ level compared to all other
136: stars within the area.
137: Stars within 120\arcsec~of one another with proper motion amplitudes
138: agreeing within 2$\sigma$ and proper motion components agreeing within
139: 1$\sigma$ in one of RA and DEC, were flagged as potential binaries.
140: We applied a color cut of
141: $z^{\prime}-J >2.5$ for at least one component to the proper-motion-selected
142: sample in order to search for BD companions. Finally all candidates
143: were examined visually to eliminate artifacts and spurious
144: matches. Of all candidate systems, G~203-50AB stood out as harboring
145: a very red companion with $z^{\prime}-J\approx2.9$.
146: Although at first glance the companion passed our color
147: cuts, the quoted $z^{\prime}$ error of $\pm2$ magnitudes rendered this color
148: meaningless.
149: Fortunately the primary was of known spectral type and absolute magnitude
150: $M_J=9.34\pm0.18$ \citep{mtcn7}, yielding $M_J=13.3$ for the
151: secondary, assuming it to be at the same distance as the primary. Average absolute magnitudes
152: as a function of spectral type \citep{dahn02} suggested that the
153: companion was indeed a mid-L dwarf.
154:
155: As a verification of the system's physical association we have plotted
156: the proper motions of G~203-50AB along with all other stars
157: within 10\arcmin~of the primary. Figure
158: \ref{fig:pm} clearly shows G~203-50AB as a co-moving pair.
159:
160: In order to establish a spectral type for the companion we obtained an
161: infrared spectrum (R$\sim$750) of 2MASS~J17114559+4028578
162: (2M1711+4028~hereafter) on 2008 Feb 28 using the
163: SpeX Medium-Resolution Spectrograph \citep{spex} at NASA's Infrared
164: Telescope Facility (IRTF). Observations were made in short wavelength
165: (0.8 to 2.5 $\mu$m) cross-dispersed mode with the $0.8\arcsec$ slit and the seeing was
166: 0.7$\arcsec$-0.9$\arcsec$. We obtained eight 180~s exposures arranged in two ABBA nod
167: patterns with a nod step of 7$\arcsec$ along the slit. For telluric and
168: instrumental transmission correction, the A0 star HD165029 was
169: observed immediately after the target at the same airmass.
170: Flat-fielding, background subtraction, spectrum extraction,
171: wavelength calibration, order merging and telluric correction were
172: done using SpeXtool \citep{spextool1,spextool2}. No scaling was
173: applied to the cross-dispersed spectrum when merging the orders. The
174: spectrum is presented and analyzed in \S\ref{sect:secondary}.
175:
176: To constrain the possible binary nature of the primary and the
177: companion, adaptive optics imaging observations of the system were
178: obtained on 2008 July 8 at the Subaru telescope (open use program
179: S08A-074). The observations were made with the AO36 adaptive optics
180: system \citep{takami03} and the CIAO near-infrared camera
181: \citep{murakawa04}. The primary star G~203-50 was used for wave front
182: sensing. Five exposures of 7~s $\times$ 3~co-additions were obtained
183: in $K_{\rm s}$ over five dither positions. The images were reduced in
184: a standard manner. A sky frame was obtained as the median of the five
185: images after masking the regions dominated by the signal from the
186: target. After subtraction of this sky frame, the images were divided
187: by a normalized dome flat image and bad pixels were replaced by a
188: median over their neighbors. All images were finally co-aligned,
189: flux-normalized, and co-added. Owing to the faint $R$-band magnitude
190: of the target star, the adaptive optics correction achieved is rather
191: poor, with a PSF FWHM of 0.18\arcsec. Visual inspection of the adaptive optics images indicates that neither the primary nor the companion
192: is a nearly equal-mass binary with a separation of $\sim$0.18\arcsec\ or larger. Subtraction of a properly shifted and scaled version of the primary star PSF from that of the companion confirms that conclusion as this operation leaves no obvious residual. Our adaptive
193: optics images also provide constraints on the existence of additional, fainter companions in the system. The detection limits achieved
194: indicate that the primary has no other companion with $\Delta K_{\rm
195: s}<4.3$, 7.0 and 7.8 mag at separation greater than 0.5\arcsec,
196: 1\arcsec\ and 2\arcsec respectively.
197: Similarly, the secondary has no companion with $\Delta K_{\rm s}<4.3$ above 0.5\arcsec.
198:
199: \section{Physical Properties}\label{sect:properties}
200: A summary of observational and physical properties of
201: the system is given in tables 1 and 2.
202: Proper motions for each component, in a reference frame defined by the
203: median proper motion of all background stars within $10\arcmin$, were
204: found and averaged to give a mean system proper motion of
205: $242\pm15$~mas~${\rm yr^{-1}}$ in right ascension and
206: $77\pm17$~mas~${\rm yr^{-1}}$ in declination. This is in good agreement with the proper
207: motion tabulated for G203-50 of $250.5\pm5.5$~mas~${\rm yr^{-1}}$ and
208: $84.2\pm5.5$~mas~${\rm yr^{-1}}$ in the Revised NLTT catalogue \citep{nltt}.
209:
210: \subsection{The primary: G 203-50} \label{sect:primary}
211: G~203-50 has been assigned a spectral type of M4.5 and an absolute
212: magnitude of $M_J=9.34\pm0.18$\footnote{errors were not provided with the
213: online data, but we have inferred them from the quoted error in
214: photometric distance} by \citet{mtcn7}. The spectral type
215: was assigned based on the TiO5 spectral index \citep{cruz02}, while the
216: absolute J magnitude is derived from polynomial fits to the TiO5, CaOH, and CaH2
217: spectral indices as a function of absolute J magnitude for M dwarfs with trigonometric
218: parallaxes \citep{cruz02}. We have verified the
219: spectral type by comparison of G 203-50's spectrum (courtesy of Neill
220: Reid) with reference spectra from \citet{kirkpatrick91} using standard
221: spectra made available online by Kelle Cruz\footnote{http://www.astro.caltech.edu/$\sim$kelle/M\_standards/}, and
222: \citet{pickles98}. The spectrum of G 203-50 is shown alongside the M4.5
223: spectral standard Gl 83.1 in figure \ref{fig:mspec}.
224:
225: Since the absolute J magnitude is derived from spectral indices which
226: may vary continuously between spectral types, it is in principle more
227: precise than a magnitude derived based on the average value for all
228: members of a given
229: subtype. However, the absolute J magnitude provided by \citet{mtcn7} seems to have
230: unrealistically low errors, given the large spread in absolute magnitude
231: for mid-M spectral types \citep[e.g. see figure 4 of ][]{cruz02}. As
232: a sanity check we conducted a SIMBAD search yielding 45 M4.5 dwarfs with
233: measured parallaxes and J band fluxes. The distribution of absolute J
234: magnitudes has a median value of 9.0, and a standard deviation of
235: 0.44. Therefore we choose to adopt a more conservative error estimate of 0.44 magnitudes.
236:
237: Using $M_J=9.34\pm0.44$, the corresponding photometric distance is
238: $22.2\pm4.5$~pc. Assuming an age $> 500$~Myr we derived a mass of $0.146\pm0.031$~\msun ~using the empirical mass-luminosity relationship of
239: \citet{delfosse2000}, also in agreement with the models of
240: \citet{baraffe98}.
241:
242: The spectrum of G 203-50 shows moderate $H\alpha$ emission. We
243: measured the $H\alpha$ equivalent width (EW) to be
244: $3.8\pm0.5$~\AA~ with errors stemming from a high sensitivity to the
245: regions chosen to fit the pseudo-continuum on either side of the
246: line. We used a cubic polynomial to fit the pseudo-continuum regions
247: from $6544.3$-$6551.9$~\AA~and from $6576.6$-$6582.3$~\AA.
248: The relationship between $H\alpha$ activity and age becomes degenerate
249: for low mass dwarfs \citep{zuckerman04} making it difficult to
250: draw firm conclusions about the age of G203-50. However,
251: the activity lifetime of M stars in terms of $H\alpha$ emission has
252: recently been constrained using a sample of 38000 M dwarfs from SDSS
253: Data Release 5 \citep{west08}. The activity lifetime for M4 and
254: M5 stars respectively is found to be 4.5 and 7~Gyr. Considering that
255: G 203-50 is more active than the majority of M dwarfs of similar
256: spectral type \citep[e.g.][see online data]{mtcn7} it is reasonable to
257: infer that it is not near the end of its active phase. Nor does
258: G 203-50 display signs of extreme youth, as it has no associated
259: x-ray source in ROSAT. Therefore we tentatively estimate the age of
260: G 203-50 to be between 1 and 5 Gyr.
261:
262: Based on the metallicity scale of \citet{gizis97} the TiO5 and CaH2
263: indices of G 203-50 are consistent with solar metallicity, indicating
264: [M/H]$>$ -1.0. We measured the metallicity index of \citet{lepine07} to
265: be $\zeta_{{\rm TiO/CaH}}=0.95$, where solar metallicity is represented by
266: $\zeta_{{\rm TiO/CaH}}=1$, and metal poor stars have $\zeta_{{\rm
267: TiO/CaH}}<0.825$. Our own fits of model spectra in the 6200-7300~\AA~
268: region verify these results. We fit the NextGen99 model atmospheres
269: \citep{hauschildt99} to G 203-50's spectrum in the region of TiO and
270: CaH bands of 6200-7300~\AA. We used a grid of models with $3000~K <
271: T_{eff} < 3300 K$ in steps of 100K, $4.5 < \log g < 5.5$ in steps of 0.5 dex, and $-2.0 < [M/H] < 0.0$
272: in steps of 0.5 dex. These parameter ranges were chosen based on
273: values of $T_{eff}=3114\pm125$ and $\log g =5.14\pm0.05$ computed from the evolutionary models of
274: \citet{baraffe98} using the 1-5 Gyr isochrones. In each case the templates were convolved with a
275: gaussian with a FWHM equal to the instrumental resolution ($\sim5.5$~\AA) of G 203-50's spectrum
276: and then interpolated onto the data. As noted by \citet{lepine07}, none of the model spectra
277: were good fits, with the CaH band consistently appearing too strong
278: relative to TiO. However, as expected the best fitting model spectra were those with solar
279: metallicity, with the fit becoming progressively worse for decreasing
280: [M/H]. Therefore we find no evidence to suggest that G 203-50 is
281: metal poor. A higher resolution spectrum is required to determine the
282: metallicity more precisely \citep[for example, using the method of][accurate to $\sim0.1$~dex]{bean06}.
283:
284:
285: \subsection{The Companion: 2MASS J17114559+4028578} \label{sect:secondary}
286: Using reference spectra from the IRTF\footnote{http://irtfweb.ifa.hawaii.edu/$\sim$spex/spexlibrary/IRTFlibrary.htm} (maintained by John Rayner), SpeX
287: Prism (maintained by Adam Burgasser), CGS4 \citep{leggett2001}, and NIRSPEC \citep{mclean03}
288: spectral libraries, we determined the best fitting spectral type for
289: the companion to be L5 (see figure \ref{fig:spectrum}). However, other reasonable
290: matches were found from L3.5 to L6.5. We also measured spectral
291: indices defined by \citet{geballe02}, \citet{tokunaga99},
292: \citet{reid01b}, and \citet{mclean03}. The indices are presented in table \ref{tbl:tab3}, yielding
293: spectral types from L4.5 to L8. We calculated synthetic 2MASS colors of $J-K_s$=1.28, $J-H$=0.79 and $H-K_s$=0.49
294: from the spectrum\footnote{the quoted
295: 2MASS $J$, $H$, and $K_s$ magnitudes are
296: flagged as being biased by the nearby primary, however our derived
297: colors agree with the former to within 0.1 mag} of 2M1711+4028 using the relative
298: spectral response curves and zero-magnitude fluxes given in
299: \citet{cohen03}. These NIR colors make 2M1711+4028 unusually blue for
300: a mid-late L dwarf \citep{cushing08}.
301: The large range in spectral type implied by the spectral indices, and
302: the unusually blue NIR colors of 2M1711+4028 likely have a common origin.
303: Several instances of anomalously blue L dwarfs have been previously
304: noted \citep[e.g.][]{knapp04,burgasser08}, where the
305: optical spectra appear normal, but the NIR colors are much bluer than
306: average. These blue L dwarfs are characterized by enhanced H2O absorption
307: bands and diminished CO, giving the appearance of a later spectral type in the NIR,
308: while J and K band features such as FeH are more consistent with an
309: earlier optical classification (see \S \ref{sect:blue} for further discussion). Similarly for 2M1711+4028, the strong H2O indices predict
310: spectral types from L6.5 to L8 while z-FeH and J-FeH indices predict types
311: $<$L6 and the K1 (not to be confused with \ion{K}{1}) index predicts a spectral type of L4.5. This is consistent with the two best matching
312: reference spectra: the relatively blue L dwarfs
313: SDSSp J05395199-0059020 \citep{fan00} and 2MASS~J15074769-1627386 \citep{reid00}, both of which
314: have optical spectral types of L5, and the latter of which is an optical spectral
315: standard. Considering all of our measurements, we have
316: assigned a NIR spectral type of $L5^{+2}_{-1.5}$ to 2M1711+4028, with our choice being most
317: strongly influenced by the best fitting reference spectra.
318: The error bars span the entire range of reasonable spectral types based on
319: template fitting and spectral indices, excluding the H2O indices.
320:
321: Using the absolute magnitude-spectral type relationship provided by
322: \citet{dahn02} an absolute magnitude of $13.5^{+0.7}_{-0.5}$ is found for
323: the secondary, corresponding to a distance of $20.0\pm6.3$~pc. This
324: is consistent with the distance of $22.2\pm4.5$~pc derived for the
325: primary. The average distance for the system is $21.2\pm3.9$~pc.
326: We find an effective temperature of $1700^{+210}_{-250}$~K based on the
327: spectral type-effective temperature relationship of
328: \citet{golimowski04}. Using the DUSTY model isochrones for 1 and 5~Gyr
329: \citep{chabrier00}, we derive a mass of
330: $0.066_{-0.015}^{+0.008}$~\msun. The quoted uncertainty takes into account the 1$\sigma$ uncertainty
331: in $ T_{eff}$ as well as the age interval. While our upper limit of
332: $0.074$~\msun~straddles the stellar-substellar boundary, this is the most
333: conservative estimate, allowing for a very broad range in spectral
334: types. We conclude that 2M1711+4028 is most likely substellar, an
335: issue which can be resolved in the future by obtaining an
336: optical spectrum in order to further constrain its spectral type.
337:
338: Based on 2MASS and SDSS astrometry 2M1711+4028 is separated from
339: G~203-50 by a mean angular separation of $6.47\arcsec\pm0.14\arcsec$, in agreement with the angular separation measured from the
340: adaptive optics images of $6.40\arcsec \pm 0.02\arcsec$ at $234.1\degr
341: \pm 0.2\degr$. This corresponds to a projected separation of $135\pm25$~AU, at the average
342: system distance.
343:
344:
345: \section{Discussion}\label{sect:discussion}
346:
347: \subsection{The blue NIR colors of 2M1711+4028}\label{sect:blue}
348: The NIR colors of mid-late L dwarfs vary significantly within a single spectral
349: type. For L5 dwarfs there is a spread of $\sim$0.7 magnitudes
350: in $J-K_s$ \citep{kirkpatrick08,cushing08}. Although surface gravity
351: and metallicity play a role, comparison of atmospheric models to
352: actual spectra \citep[e.g.][]{knapp04, cushing08,
353: burgasser08} suggests that large variations in the NIR colors of L
354: dwarfs are primarily related to the properties of condensate clouds in their
355: atmospheres, with unusually red SEDs arising from thick clouds and
356: blue ones from thin or large-grained clouds.
357: Common to the known peculiar blue L dwarfs is exaggerated H2O
358: absorption and diminished CO, as seen in the spectrum of 2M1711+4028.
359: As discussed in \S \ref{sect:secondary} the discrepancy between the
360: late-type H20 indices and the earlier type FeH and K1 indices, along
361: with its unusually blue NIR colors are indications that 2M1711+4028
362: falls into this category. For comparison we overplot 2M1711+4028's spectrum with the
363: spectra of the very red L4.5 dwarf 2MASS J22244381-0158521 \citep{kirkpatrick00}, and the
364: relatively blue L5 optical standard 2MASS~J15074769-1627386 (see figure \ref{fig:blue2}). All spectra
365: agree reasonably well in the J-band but diverge significantly at H and K, which may
366: indicate differing properties of condensate clouds in their atmospheres.
367: As a member of a wide binary system the surface gravity and
368: metallicity of 2M1711+4028 can be constrained from properties of the
369: primary, yielding an excellent laboratory for studying BD
370: atmospheres. The estimated age of 1-5~Gyr for the G 203-50 primary
371: implies a relatively high surface gravity for the companion, possibly
372: contributing the its blue NIR colors. However
373: surface gravity alone does not seem to be sufficient in explaining the
374: NIR colors of peculiar blue L
375: dwarfs \citep{burgasser08}. Additionally, the
376: primary shows no signs of being metal poor, lending support to the
377: hypothesis that unusually blue NIR colors can be primarily attributed
378: to cloud properties. Higher resolution spectroscopy of the primary
379: and a more precise determination of the metallicity is required in
380: order to confirm this conclusion.
381:
382: Another potential cause of unusually blue NIR colors is unresolved
383: binarity. This may explain the slight onset of CH4 absorption at 2.2 $\mu$m in
384: the spectrum of 2M1711+4028. At least one of the known peculiar blue L dwarfs, 2MASS
385: J08053189+4812330 \citep{burgasser07b} is thought to be an unresolved
386: binary with L4.5 and T5 components. However, this system exhibits a
387: pronounced dip in the 1.6 $\mu$m CH4 feature due to the peaked shape of
388: the T dwarf's SED, whereas the spectrum of 2M1711+4028 is
389: relatively flat in that region. Furthermore, the SED and colors of our
390: companion are very similar to that of another blue L4.5 dwarf, 2MASS
391: J11263991-5003550 (2M1126-5003 hereafter), discussed at length by
392: \citet{burgasser08}. By constructing composite spectra using published L and T dwarf spectra from the SpeX
393: prism library, \citet{burgasser08} determined that no reasonable
394: composite spectrum could be found that matched that of 2M1126-5003 in
395: both the optical and NIR. Without an optical spectrum for 2M1711+4028
396: we are limited in the conclusions we can draw, but its similarities to
397: 2M1126-5003 may suggest that 2M1711+4028 is a single BD.
398: Our AO images support this conclusion, indicating that the BD is not a
399: near-equal mass binary with separation $> 0.18\arcsec$.
400:
401: \subsection{Formation of G 203-50AB}\label{sect:formation}
402:
403: With a total mass of $\sim$0.21~\msun~G~203-50AB is slightly more massive than the rare
404: wide VLM binaries, but much less massive than the solar analogues
405: around which BDs are routinely found at wide separations (see \S
406: \ref{sect:intro}, and figure \ref{fig:sep}). It is
407: therefore of interest to consider how G~203-50AB may have
408: formed. Could the secondary have formed through
409: gravitational instability in a disk around the primary? Given the mass ratio
410: of $q$=0.45, that would imply a $M_{disk} >$ 0.45$M_*$, whereas typical
411: disks around low-mass stars contain a few percent of $M_*$ at $\sim$1 Myr \citep{scholz06}.
412: Since it is unlikely that the entire disk would end up in the companion, the
413: total disk mass, even in a conservative estimate, would have to be larger
414: than the primary's own mass to start with. Thus, we conclude that formation
415: of 2M1711+4028 in a protostellar disk around G~203-50 is implausible.
416:
417: On the other hand, gravitational fragmentation of prestellar cores appears
418: to be capable of forming a wide variety of binary systems, depending on the
419: size, mass and angular momentum of the core \citep[e.g.][]{bate00}. However,
420: simulations usually have some difficulty producing binary stars with low
421: component masses and wide separations \citep[e.g.][]{bate03,goodwin04}. Some theoretical models invoke ejection from the parent cloud to
422: halt further accretion that would otherwise lead to higher masses. Given its
423: projected separation of 135$\pm$25~AU, the G~203-50AB binary has a binding
424: energy of 12.6$\pm3.8~\times$~$10^{-41}$~erg, placing it below the empirical
425: ``minimum'' noted by \citet{close03} and \citet{burgasser07a}. Thus, it is
426: unlikely to have survived such an
427: ejection. We suggest that G~203-50AB most likely formed via fragmentation of an isolated core and
428: did not suffer strong dynamical interactions during the birth process or
429: subsequently.
430:
431:
432: \subsection{Search Sensitivity}\label{sect:sensitivity}
433: In order to assess the sensitivity of our search we simulated proper
434: motion distributions of M dwarfs distributed uniformly in a spherical volume out
435: to 25 pc, with tangential space velocities drawn from the distribution
436: of \cite{schmidt07}. To be sensitive
437: to a particular M dwarf primary, its displacement between the 2MASS and
438: SDSS surveys had to be greater than the $3\sigma$ dispersion of all
439: other stars in the 4~${\rm deg^2}$ section of the sky in which it was
440: found. For each such section of the sky, the time baseline between the
441: surveys was computed and used to determine the minimum proper motion
442: required for a detection. We assumed that the population of M dwarfs
443: within 25 pc was uniformly distributed and assigned equal weight to
444: each 4 ${\rm deg^2}$ area of the sky. Using the simulated proper motion
445: distributions the fraction of M dwarfs whose proper motions we could
446: have measured was determined for each section of the sky, giving an
447: average fraction of 0.58. Adopting an M dwarf space
448: density ($9.0 < M_V < 13.0$ or roughly M0.5-M5.5) of $283.37
449: \times10^{-4}~{\rm pc^{-3}}$ \citep{reid02}, and given a search area of 7668
450: ${\rm deg^2}$ of the sky, we should have been sensitive to approximately
451: $201\pm12$ early-mid M dwarfs within 25 pc.
452:
453: Although in some cases we were able to recover binary systems with
454: separations $<$~$4\arcsec$, we conservatively put a lower limit of
455: $6\arcsec$ on our sensitivity, ensuring that components are well
456: separated. The upper limit for separation is set by our search
457: radius, which extended to $120\arcsec$. These limits correspond to
458: projected separations of 30-600~AU at 5~pc and 150-3000~AU at 25~pc. Our sensitivity to
459: companions around each star was dictated by the mean 2MASS J-band
460: limiting magnitude (S/N=10) of $\sim16.5$, corresponding to a minimum mass of
461: $\sim0.06$~\msun~at 25~pc, assuming an age of 1-5 Gyr. Other factors preventing us from finding
462: companions include poor astrometry due to saturation of the primary,
463: or low S/N of the secondary. To estimate the number of
464: binaries missed we used SIMBAD and
465: DwarfArchives\footnote{http://DwarfArchives.org} to compile a list of
466: 31 M-dwarfs and 38 BDs with previously measured proper motions large
467: enough to pass our cuts, and tested whether we
468: could measure the same proper motions using SDSS and 2MASS astrometry.
469: We found that $91\%$ of the time for M dwarfs, and $79\%$ of
470: the time for BDs our measured proper motions agreed with the
471: previously measured ones, using the same criteria as our matching
472: algorithm described in \S\ref{sect:search}. Therefore, we should have
473: been capable of identifying approximately $72\%$ of binaries
474: with sufficiently high proper motions. Correspondingly we adjust our
475: sensitivity to $\sim145\pm9$ M dwarfs. Adopting Poisson uncertainties on
476: a $1\sigma$ confidence interval for our single detection we roughly estimate that $0.7^{+1.5}_{-0.6}\%$
477: of early-mid M dwarfs have substellar companions with masses greater
478: than $\sim0.06$~\msun, at separations above $\sim120$~AU.
479:
480: \section{Summary and Outlook}\label{sect:conclusions}
481:
482: Above we have outlined our discovery of a wide, unusually blue, L5 companion to the nearby
483: M4.5 dwarf G 203-50. Since BDs cool with time, it is not possible to infer their masses from observed luminosities. In order to break this degeneracy, the age (or
484: an age indicator such as gravity) of the BD must be known. Even so,
485: determining the mass of a BD requires accurate evolutionary models.
486: In order to constrain these models we must rely on a handful of BDs
487: for which independent age estimates can be obtained. Our companion, 2M1711+4028 falls into this category, as its age can be constrained from the
488: age of the primary. With
489: an angular separation of over 6\arcsec ~the components of G~203-50AB
490: are well separated, allowing the primary and secondary to be studied
491: independently. At a distance of only $\sim21$~pc, the parallax can be
492: measured relatively easily, providing a more precise
493: determination of distance and luminosity. Assuming an age
494: of 1-5 Gyr, 2M1711+4028 is older than most BDs with
495: independent age estimates (e.g. those found in star forming regions)
496: and can therefore provide an anchor point in a poorly constrained
497: mass-age regime. Furthermore, as an unusually blue L dwarf in the NIR, 2M1711+4028
498: provides a unique opportunity for studying the relative importance of
499: gravity, metallicity and cloud properties in determining the NIR
500: colors of L dwarfs.
501:
502: With a total mass falling between the solar mass and VLM
503: regimes, G~203-50AB also has an important bearing on star formation theory. Based
504: on the large mass ratio between the system components, we rule out
505: formation of the companion in the disk of the primary. Instead we
506: suggest that this weakly bound binary formed via the fragmentation of
507: an isolated core, and did not suffer disruptive dynamical
508: interactions. Statistically we put an upper limit of $2.2\%$ on the
509: wide companion fraction for BD companions with $m > 0.06$~\msun ~around
510: early-to-mid M dwarfs.
511:
512: In order to better constrain the properties of this unique system we
513: recommend that future observations of G~203-50AB include: a
514: parallax measurement to resolve uncertainties over the distance and
515: absolute magnitude of G~203-50; a high resolution spectrum of G~203-50
516: in order to determine the metallicity more precisely; an optical
517: spectrum for 2M1711+4028 to determine an optical spectral type; and time
518: series spectra to check for spectroscopic binarity.
519:
520: \acknowledgments
521: We thank the anonymous referee for a prompt and
522: thoughtful review that greatly improved the quality of this manuscript. We would like to thank the IRTF support staff, especially John Rayner
523: for walking us through SpeX observing procedures, and sharing with us
524: his valuable knowledge of the instrument. We would
525: also like to thank Neil Reid for providing us with a spectrum for G
526: 203-50. This research has benefited
527: from the SIMBAD database,operated at CDS, Strasbourg, France; the M, L and T dwarf compendium
528: housed at DwarfArchives.org and maintained by Chris Gelino, Davy Kirkpatrick, and
529: Adam Burgasser; The VLM Binaries Archive at VLMBinaries.org,
530: maintained by Nick Siegler; The SpeX Prism Spectral Libraries, maintained by Adam
531: Burgasser; the IRTF Spectral Library maintained by John
532: Rayner; and the M dwarf standard spectra
533: made available by Kelle Cruz at
534: http://www.astro.caltech.edu/$\sim$kelle/M\_standards. This publication makes use of data products
535: from 2MASS, which is a joint project of the University of
536: Massachusetts and the Infrared Processing and Analysis
537: Center/California Institute of Technology, funded by the National
538: Aeronautics and Space Administration and the National Science
539: Foundation. This publication also makes use of data products from SDSS, which is funded by the Alfred P. Sloan Foundation, the Participating
540: Institutions, the National Science Foundation, the U.S. Department of
541: Energy, the National Aeronautics and Space Administration, the
542: Japanese Monbukagakusho, the Max Planck Society, and the Higher
543: Education Funding Council for England. Observations were based in part on data collected at Subaru Telescope, which is operated by the National Astronomical Observatory of Japan. JR
544: is supported in part by an Ontario Graduate Scholarship. DL is
545: supported in part through a postdoctoral fellowship from the Fonds Qu\'eb\'ecois de la Recherche
546: sur la Nature et les Technologies. This work was supported in part
547: through grants to RJ and RD from the Natural Sciences and Engineering
548: Research Council (NSERC), Canada, and an Early Researcher Award from
549: the province of Ontario to RJ.
550:
551: \begin{thebibliography}
552: {}
553: \expandafter\ifx\csname natexlab\endcsname\relax\def\natexlab#1{#1}\fi
554:
555: \bibitem[Adelman-McCarthy
556: \& et al.(2008)]{sdss} Adelman-McCarthy, J.~K., \& et
557: al.\ 2008, VizieR Online Data Catalog, 2282, 0
558: \bibitem[Allen \& Reid(2008)]{allen08} Allen, P.~R., \& Reid, I.~N.\ 2008, ArXiv e-prints, 804, arXiv:0804.2872
559: \bibitem[Baraffe et al.(1998)]{baraffe98} Baraffe, I., Chabrier, G., Allard, F., \& Hauschildt, P.~H.\ 1998, \aap, 337, 403
560: \bibitem[Bate(2000)]{bate00} Bate, M.~R.\ 2000, \mnras, 314,
561: 33
562: \bibitem[Bate et al.(2003)]{bate03} Bate, M.~R., Bonnell,
563: I.~A., \& Bromm, V.\ 2003, \mnras, 339, 577
564: \bibitem[Bate
565: \& Bonnell(2005)]{bate05} Bate, M.~R., \& Bonnell, I.~A.\ 2005, \mnras, 356, 1201
566: \bibitem[Bean et al.(2006)]{bean06} Bean, J.~L., Sneden, C.,
567: Hauschildt, P.~H., Johns-Krull, C.~M.,
568: \& Benedict, G.~F.\ 2006, \apj, 652, 1604
569: \bibitem[Boss(2000)]{boss00} Boss, A.~P.\ 2000, \apjl, 536,
570: L101
571: \bibitem[Burgasser et al.(2008)]{burgasser08} Burgasser, A.~J.,
572: Looper, D.~L., Kirkpatrick, J.~D., Cruz, K.~L.,
573: \& Swift, B.~J.\ 2008, \apj, 674, 451
574: \bibitem[Burgasser et al.(2007a)]{burgasser07a} Burgasser, A.~J.,Reid, I.~N., Siegler, N., Close, L., Allen, P., Lowrance, P., \& Gizis, J.\
575: 2007a, Protostars and Planets V, 427
576: \bibitem[Burgasser(2007b)]{burgasser07b} Burgasser, A.~J.\ 2007b, \aj,
577: 134, 1330
578: \bibitem[Burgasser et al.(2005)]{burgasser05} Burgasser, A.~J., Kirkpatrick, J.~D., \& Lowrance, P.~J.\ 2005, \aj, 129, 2849
579: \bibitem[Chabrier et al.(2000)]{chabrier00} Chabrier, G., Baraffe,
580: I., Allard, F., \& Hauschildt, P.\ 2000, \apj, 542, 464
581: \bibitem[Cohen et al.(2003)]{cohen03} Cohen, M., Wheaton,
582: W.~A., \& Megeath, S.~T.\ 2003, \aj, 126, 1090
583: \bibitem[Cruz~\&~Reid(2002)]{cruz02} Cruz, K.~L., \& Reid, I.~N.\ 2002, \aj, 123, 2828
584: \bibitem[Close et al.(2003)]{close03} Close, L.~M., Siegler, N., Freed, M., \& Biller, B.\ 2003, \apj, 587, 407
585: \bibitem[Close et al.(1990)]{close90} Close, L.~M., Richer,
586: H.~B., \& Crabtree, D.~R.\ 1990, \aj, 100, 1968
587: \bibitem[Cushing et al.(2008)]{cushing08} Cushing, M.~C., et al.\
588: 2008, \apj, 678, 1372
589: \bibitem[Cushing et al.(2004)]{spextool1} Cushing, M.~C., Vacca, W.~D., \& Rayner, J.~T.\ 2004, \pasp, 116, 362
590: \bibitem[Cushing et al.(2005)]{cushing05} Cushing, M.~C., Rayner,
591: J.~T., \& Vacca, W.~D.\ 2005, \apj, 623, 1115
592: \bibitem[Daemgen et al.(2007)]{daemgen07} Daemgen, S., Siegler, N., Reid, I.~N., \& Close, L.~M.\ 2007, \apj, 654, 558
593: \bibitem[Dahn et al.(2002)]{dahn02} Dahn, C.~C., et al.\ 2002,
594: \aj, 124, 1170
595: \bibitem[Delfosse et al.(2000)]{delfosse2000} Delfosse, X., Forveille, T., S{\'e}gransan, D., Beuzit, J.-L., Udry, S., Perrier, C., \& Mayor, M.\ 2000, \aap, 364, 217
596: %\bibitem[Delfosse et al.(2004)]{delfosse04} Delfosse, X., et al.\
597: %2004, Spectroscopically and Spatially Resolving the Components of the Close
598: %Binary Stars, 318, 166
599: %\bibitem[Duquennoy
600: %\& Mayor(1991)]{duquennoy01} Duquennoy, A., \& Mayor, M.\ 1991, \aap, 248, 485
601: \bibitem[Fan et al.(2000)]{fan00} Fan, X., et al.\ 2000, \aj,
602: 119, 928
603: \bibitem[Fischer
604: \& Marcy(1992)]{FM92} Fischer, D.~A., \& Marcy, G.~W.\ 1992, \apj, 396, 178
605: \bibitem[Geballe et al.(2002)]{geballe02} Geballe, T.~R., et al.\
606: 2002, \apj, 564, 466
607: \bibitem[Gizis(1997)]{gizis97} Gizis, J.~E.\ 1997, \aj, 113,
608: 806
609: \bibitem[Goldman et al.(1999)]{goldman99} Goldman et al.\ 1999, \aap, 351, L5
610: \bibitem[Golimowski et al.(2004)]{golimowski04} Golimowski, D.~A.,
611: et al.\ 2004, \aj, 127, 3516
612: \bibitem[Goodwin et al.(2004)]{goodwin04} Goodwin, S.~P., Whitworth, A.~P., \& Ward-Thompson, D.\ 2004, \aap, 414, 633
613: %\bibitem[Hawley et al.(2002)]{hawley02} Hawley, S.~L., et al.\
614: %2002, \aj, 123, 3409
615: \bibitem[Hauschildt et al.(1999)]{hauschildt99} Hauschildt, P.~H.,
616: Allard, F., \& Baron, E.\ 1999, \apj, 512, 377
617: \bibitem[Hinz et al.(2002)]{hinz02} Hinz, J.~L., McCarthy,
618: D.~W., Jr., Simons, D.~A., Henry, T.~J., Kirkpatrick, J.~D.,
619: \& McGuire, P.~C.\ 2002, \aj, 123, 2027
620: Simons, D.~A., Henry, T.~J., Kirkpatrick, J.~D.,
621: \bibitem[Kirkpatrick(2008)]{kirkpatrick08} Kirkpatrick, J.~D.\ 2008,
622: 14th Cambridge Workshop on Cool Stars, Stellar Systems, and the Sun, 384,
623: 85
624: \bibitem[Kirkpatrick et al.(2000)]{kirkpatrick00} Kirkpatrick, J.~D.,
625: et al.\ 2000, \aj, 120, 447
626: \bibitem[Kirkpatrick, Henry, \& McCarthy(1991)]{kirkpatrick91} Kirkpatrick, J.~D., Henry, T.~J., \& McCarthy, D.~W., Jr.\ 1991, \apjs, 77, 417
627: \bibitem[Knapp et al.(2004)]{knapp04} Knapp, G.~R., et al.\
628: 2004, \aj, 127, 3553
629: \bibitem[L{\'e}pine et al.(2007)]{lepine07} L{\'e}pine, S.,
630: Rich, R.~M., \& Shara, M.~M.\ 2007, \apj, 669, 1235
631: \bibitem[Leggett et al.(2001)]{leggett2001} Leggett, S.~K., Allard,
632: F., Geballe, T.~R., Hauschildt, P.~H.,
633: \& Schweitzer, A.\ 2001, \apj, 548, 908
634: \bibitem[Lowrance et al.(2000)]{lowrance99} Lowrance, P.~J., et
635: al.\ 2000, \apj, 541, 390
636: \bibitem[Marcy
637: \& Butler(2000)]{marcy00} Marcy, G.~W., \& Butler, R.~P.\ 2000, \pasp, 112, 137
638: %\bibitem[Mason et al.(1998)]{mason98} Mason, B.~D., Gies, D.~R., Hartkopf, W.~I., Bagnuolo, W.~G., Jr., ten Brummelaar, T.,
639: %\& McAlister, H.~A.\ 1998, \aj, 115, 821 %O-star binary fraction > 75%
640: \bibitem[McCarthy~\&~Zuckerman(2004)]{mccarthy04} McCarthy, C., \& Zuckerman, B.\ 2004, \aj, 127, 2871
641: \bibitem[McLean et al.(2003)]{mclean03} McLean, I.~S., McGovern,
642: M.~R., Burgasser, A.~J., Kirkpatrick, J.~D., Prato, L.,
643: \& Kim, S.~S.\ 2003, \apj, 596, 561
644: \bibitem[Metchev~\& Hillenbrand(2005)]{met05} Metchev, S.~A., \& Hillenbrand, L.~A.\ 2005, Protostars and Planets V, 8635
645: \bibitem[Murakawa et al.(2004)]{murakawa04} Murakawa, K., et al.\ 2004, \pasj, 56, 509
646: \bibitem[Nakajima et al.(1995)]{nakajima95} Nakajima, T.,
647: Oppenheimer, B.~R., Kulkarni, S.~R., Golimowski, D.~A., Matthews, K.,
648: \& Durrance, S.~T.\ 1995, \nat, 378, 463
649: \bibitem[Oppenheimer et al.(2001)]{oppen01} Oppenheimer, B.~R., Golimowski, D.~A., Kulkarni, S.~R., Matthews, K., Nakajima, T.,
650: Creech-Eakman, M., \& Durrance, S.~T.\ 2001, \aj, 121, 2189
651: \bibitem[Pickles(1998)]{pickles98} Pickles, A.~J.\ 1998, \pasp,
652: 110, 863
653: \bibitem[Padoan \& Nordlund(2002)]{padoan02} Padoan, P., \&
654: Nordlund, {\AA}.\ 2002, \apj, 576, 870
655:
656: \bibitem[Pinfield et al.(2006)]{pinfield06} Pinfield, D.~J.,
657: Jones, H.~R.~A., Lucas, P.~W., Kendall, T.~R., Folkes, S.~L., Day-Jones,
658: A.~C., Chappelle, R.~J., \& Steele, I.~A.\ 2006, \mnras, 368, 1281
659: \bibitem[Rayner et al.(2003)]{spex} Rayner, J.~T., Toomey, D.~W., Onaka, P.~M., Denault, A.~J., Stahlberger, W.~E., Vacca, W.~D.,
660: Cushing, M.~C., \& Wang, S.\ 2003, \pasp, 115, 362 %spex paper
661: \bibitem[Rebolo et al.(1998)]{rebolo98} Rebolo, R., Zapatero
662: Osorio, M.~R., Madruga, S., Bejar, V.~J.~S., Arribas, S.,
663: \& Licandro, J.\ 1998, Science, 282, 1309
664: \bibitem[Reid
665: \& Walkowicz(2006)]{reid06} Reid, I.~N., \& Walkowicz, L.~M.\ 2006, \pasp, 118, 671
666: \bibitem[Reid et al.(2000)]{reid00} Reid, I.~N., Kirkpatrick,
667: J.~D., Gizis, J.~E., Dahn, C.~C., Monet, D.~G., Williams, R.~J., Liebert,
668: J., \& Burgasser, A.~J.\ 2000, \aj, 119, 369
669: \bibitem[Reid et al.(2001a)]{reid01} Reid, I.~N., Gizis, J.~E.,
670: Kirkpatrick, J.~D., \& Koerner, D.~W.\ 2001a, \aj, 121, 489
671: \bibitem[Reid et al.(2001b)]{reid01b} Reid, I.~N., Burgasser,
672: A.~J., Cruz, K.~L., Kirkpatrick, J.~D.,
673: \& Gizis, J.~E.\ 2001b, \aj, 121, 1710
674: \bibitem[Reid et al.(2002a)]{reid02} Reid, I.~N., Gizis, J.~E.,\& Hawley, S.~L.\ 2002a, \aj, 124, 2721
675: \bibitem[Reid et al.(2003)]{mtcn7} Reid, I.~N., et al.\ 2003, \aj, 126, 3007 %Meeting the cool neighbors spectroscopy of NLTT dwarfs
676: \bibitem[Reid~\& Cruz(2002b)]{reid02b} Reid, I.~N., \& Cruz, K.~L.\ 2002b, \aj, 123, 2806
677: \bibitem[Reid
678: \& Gizis(1997)]{reid97b} Reid, I.~N., \& Gizis, J.~E.\ 1997, \aj, 114, 1992
679: \bibitem[Reipurth \& Clarke(2001)]{rep01} Reipurth, B., \&
680: Clarke, C.\ 2001, \aj, 122, 432
681: \bibitem[Salim \& Gould(2003)]{nltt} Salim, S., \& Gould, A.\ 2003, \apj, 582, 1011 %NLTT catalogue
682: \bibitem[Schmidt et al.(2007)]{schmidt07} Schmidt, S.~J., Cruz, K.~L., Bongiorno, B.~J., Liebert, J., \& Reid, I.~N.\ 2007, \aj, 133, 2258
683: \bibitem[Scholz et al.(2006)]{scholz06} Scholz, A.,
684: Jayawardhana, R., \& Wood, K.\ 2006, \apj, 645, 1498
685: \bibitem[Skrutskie et al.(2006)]{skrutskie06} Skrutskie, M.~F., et
686: al.\ 2006, \aj, 131, 1163
687: \bibitem[Stamatellos et al.(2007)]{stamatellos07} Stamatellos, D.,
688: Hubber, D.~A., \& Whitworth, A.~P.\ 2007, \mnras, 382, L30
689: \bibitem[Takami et al.(2004)]{takami03} Takami, H., et al.\
690: 2004, \pasj, 56, 225
691: \bibitem[Tokovinin(1997)]{tokovinin97} Tokovinin, A.~A.\ 1997, \aaps, 124, 75
692: \bibitem[Tokunaga
693: \& Kobayashi(1999)]{tokunaga99} Tokunaga, A.~T., \& Kobayashi, N.\ 1999, \aj, 117, 1010
694: \bibitem[Vacca et al.(2003)]{spextool2} Vacca, W.~D., Cushing, M.~C.,
695: \& Rayner, J.~T.\ 2003, \pasp, 115, 389
696: %\bibitem[Watkins et al.(1998)]{watkins98} Watkins, S.~J.,
697: %Bhattal, A.~S., Boffin, H.~M.~J., Francis, N.,
698: %\& Whitworth, A.~P.\ 1998, \mnras, 300, 1214
699: \bibitem[West et al.(2008)]{west08} West, A.~A., Hawley,
700: S.~L., Bochanski, J.~J., Covey, K.~R., Reid, I.~N., Dhital, S., Hilton,
701: E.~J., \& Masuda, M.\ 2008, \aj, 135, 785
702: \bibitem[Wilson et al.(2001)]{wilson01} Wilson, J.~C.,
703: Kirkpatrick, J.~D., Gizis, J.~E., Skrutskie, M.~F., Monet, D.~G.,
704: \& Houck, J.~R.\ 2001, \aj, 122, 1989
705: \bibitem[Zuckerman \& Song(2004)]{zuckerman04} Zuckerman, B., \& Song, I.\ 2004, \araa, 42, 685 %young stars near the sun
706: \end{thebibliography}
707:
708: \clearpage
709:
710: \input{tab1}
711:
712: \clearpage
713:
714: \input{tab2}
715:
716: \clearpage
717:
718: \input{tab3}
719:
720:
721: \clearpage
722:
723: \begin{figure}[here]
724: \epsscale{1}
725: \plotone{f1.eps}
726: \caption{Proper motions of all sources within $10\arcmin$ of G~203-50. Circles indicate the $1\sigma$
727: and $3\sigma$ dispersions. G203-50 and 2M1711+4028 are displayed
728: with error bars based on the $1\sigma$ dispersions along each axis,
729: which dominates over the astrometric errors quoted by the individual
730: catalogues. 2MASS and SDSS observations of this system are separated by a baseline of 6.0959 years.
731: \label{fig:pm}}
732: \end{figure}
733:
734: \clearpage
735:
736: \begin{figure}[here]
737: \epsscale{1}
738: \plotone{f2.eps}
739: \caption{Spectrum of G 203-50 \citep{mtcn7}, plotted alongside
740: the M4.5 spectral standard Gl 83.1 for comparison.
741: \label{fig:mspec}}
742: \end{figure}
743:
744: \clearpage
745: \begin{figure}[here]
746: \epsscale{1}
747: \plotone{f3.eps}
748: \caption{Spectrum of 2M1711+4028 obtained at IRTF using the SpeX
749: Medium-Resolution Spectrograph. The spectrum is plotted alongside
750: the L5 dwarf SDSSp~J05395199-0059020 \citep{fan00}, from the IRTF Spectral
751: Library. Both spectra are normalized in 1.27-1.33~$\mu$m and offset for clarity.
752: \label{fig:spectrum}}
753: \end{figure}
754:
755: \clearpage
756:
757: \begin{figure}[here]
758: \epsscale{1}
759: \plotone{f4.eps}
760: \caption{Spectrum of blue ($J-K_s=1.28$) L dwarf 2M1711+4028 (solid
761: line, this paper) overplotted with spectrum
762: of relatively red ($J-K_s=2.05$) L4.5 dwarf 2MASS J22244381-0158521
763: \citep[top dotted line,][]{kirkpatrick00}
764: and relatively blue ($J-K_S=1.52$) L5 optical standard 2MASS
765: J15074769-1627386 \citep[dashed line,][]{reid00}. Both comparison spectra are from the IRTF Spectral Library.
766: All spectra are normalized in 1.27-1.29~$\mu$m. They have been rebinned to a
767: lower resolution for clarity.
768: \label{fig:blue2}}
769: \end{figure}
770:
771: \clearpage
772:
773: \begin{figure}[here]
774: \epsscale{1}
775: \plotone{f5.eps}
776: \caption{Separation versus total system mass for known binary systems.
777: Stellar binaries (dots) are from \citet{tokovinin97,FM92,reid97b,close90}; VLM
778: binaries (open squares) are from the VLM binary archive, maintained by Nick
779: Siegler (vlmbinaries.org); BD-stellar binaries (filled circles) are
780: from \citet{reid01,met05}. G~203-50AB (star symbol) falls in
781: between the VLM and solar analogue regimes, and appears to be more
782: loosely bound than most systems of similar mass. BD-stellar
783: systems with M-dwarf primaries from \citet{reid01} have been
784: circled. The large triangle and square are BD-M dwarf systems from
785: \citet{wilson01} and \citet{reid06}.
786: \label{fig:sep}}
787: \end{figure}
788:
789:
790: \end{document}
791: