1:
2: %% The command below calls the default manuscript style,
3: %% which will produce a double-spaced document on one column.
4: % \documentclass{aastex,epsfig}
5:
6: %% preprint produces a one-column, single-spaced document:
7: %\documentclass[12pt,preprint]{aastex}
8: \documentclass{emulateapj}
9: %\include{epsfig.sty}
10: %% preprint2 produces a double-column, single-spaced document:
11: % \documentclass[preprint2]{aastex}
12:
13: \citestyle{aa}
14: \newcommand{\vdag}{(v)^\dagger}
15: \newcommand{\myemail}{jmonnier@cfa.harvard.edu}
16: \newcommand {\JDM}{\bf \Large}
17: \newcommand{\tmin}{T$_{\rm min}$}
18: \newcommand{\tmax}{T$_{\rm max}$}
19: \newcommand{\ee}[2]{#1 \times 10^{#2}}
20: \newcommand{\K}{\,{\rm K}}
21: \newcommand{\secd}{\,{\rm s}}
22: \newcommand{\km}{\,{\rm km}}
23: \newcommand{\yr}{\,{\rm yr}}
24: \newcommand{\masyr}{\,{\rm mas}\yr^{-1}}
25: \newcommand{\simle}{\mbox{$\stackrel{<}{_{\sim}}$}}
26: \newcommand{\simge}{\mbox{$\stackrel{>}{_{\sim}}$}}
27: %\newcommand\arcdeg{\hbox{$^\circ$} }
28: \newcommand\arcdegg{\hbox{$^\circ$}}
29: \newcommand{\jmicron}{\,\hbox{$\mu$m} }
30: \newcommand{\micronn}{\,\hbox{$\mu$m}}
31: \newcommand\etal{ {\em et~al.\/}\thinspace}
32: %\newcommand\arcsec{\hbox{$^{\prime\prime}$} }
33: \newcommand\arcsecc{\hbox{$^{\prime\prime}$}}
34: \newcommand\msun{\hbox{\,M$_\odot$}}
35: \newcommand\rsun{\hbox{\,R$_\odot$}}
36: \newcommand\lsun{\hbox{\,L$_\odot$}}
37: \newcommand\kms{ km~s$^{-1}$}
38: \newcommand\rstar{\,R$_\star$}
39: \newcommand\wn{\,cm$^{-1}$}
40:
41: %\renewcommand{\thefootnote}{\fnsymbol{footnote}}
42:
43:
44:
45: %% You can insert a short comment on the title page using the command below.
46: \slugcomment{Accepted for publication in ApJ Letters}
47:
48: %% If you wish, you may supply running head information, although
49: %% this information may be modified by the editorial offices.
50: %% The left head contains a list of authors,
51: %% usually a maximum of three (otherwise use et al.). The right
52: %% head is a modified title of up to roughly 44 characters. Running heads
53: %% will not print in the manuscript style.
54: \shorttitle{The Circumbinary Disk of v892 Tau}
55: \shortauthors{Monnier et al.}
56:
57: \begin{document}
58:
59: %% LaTeX will automatically break titles if they run longer than
60: %% one line. However, you may use \\ to force a line break if
61: %% you desire.
62:
63: \title{Discovery of a Circumbinary Disk around Herbig Ae/Be system v892~Tau}
64:
65: %% Use \author, \affil, and the \and command to format
66: %% author and affiliation information.
67: %% Note that \email has replaced the old \authoremail command
68: %% from AASTeX v4.0. You can use \email to mark an email address
69: %% anywhere in the paper, not just in the front matter.
70: %% As in the title, you can use \\ to force line breaks.
71:
72: \author{J.~D.~Monnier\altaffilmark{1},
73: A. Tannirkulam\altaffilmark{1},
74: P. G. Tuthill\altaffilmark{2},
75: M. Ireland\altaffilmark{2}, R. Cohen\altaffilmark{3},
76: W. C. Danchi\altaffilmark{4},
77: F. Baron\altaffilmark{5}
78: }
79:
80: \altaffiltext{1}{monnier@umich.edu: University of Michigan Astronomy Department, 941 Dennison Bldg, Ann Arbor, MI 48109-1090, USA.}
81: \altaffiltext{2}{University of Sydney}
82: %\altaffiltext{3}{California Institute of Technology}
83: \altaffiltext{3}{W. M. Keck Observatory}
84: \altaffiltext{4}{NASA-GSFC}
85: \altaffiltext{5}{University of Cambridge}
86: \email{JDM: monnier@umich.edu}
87:
88: %% Mark off your abstract in the ``abstract'' environment. In the manuscript
89: %% style, abstract will output a Received/Accepted line after the
90: %% title and affiliation information. No date will appear since the author
91: %% does not have this information. The dates will be filled in by the
92: %% editorial office after submission.
93:
94: \begin{abstract}
95: We report the discovery of a circumbinary disk around the Herbig
96: Ae/Be system v892~Tau. Our detailed mid-infrared images were made
97: using segment-tilting interferometry on the Keck-1 Telescope and
98: reveal an asymmetric disk inclined at $\sim$60$\arcdeg$ with an
99: inner hole diameter of 250~mas (35~AU), approximately 5$\times$
100: larger than the apparent separation of the binary components. In
101: addition, we report a new measurement along the binary orbit using
102: near-infrared Keck aperture masking, allowing a crude estimate of
103: orbital parameters and the system mass for the first time. The size
104: of the inner hole appears to be consistent with the minimum size
105: prediction from tidal truncation theory, bearing a resemblance to
106: the recently unmasked binary CoKu Tau/4. Our results have motivated a
107: re-analysis of the system spectral energy distribution, concluding
108: the luminosity of this system has been severely underestimated.
109: With further study and monitoring, v892~Tau should prove a
110: powerful testing ground for both predictions of dynamical models for
111: disk-star interactions in young systems with gas-rich disks and for
112: calibrations of pre-main-sequence tracks for intermediate-mass stars.
113:
114: %the dust distribution along the colliding wind interface can
115: %be reconstructed from these infrared images,
116:
117: \end{abstract}
118:
119: %% Keywords should appear after the \end{abstract} command. The uncommented
120: %% example has been keyed in ApJ style. See the instructions to authors
121: %% for the journal to which you are submitting your paper to determine
122: %% what keyword punctuation is appropriate.
123:
124: \keywords{techniques:interferometric, stars: pre--main sequence, stars: binaries: close, infrared:stars, stars: individual (v892~Tau)
125: }
126:
127: %% From the front matter, we move on to the body of the paper.
128: %% In the first two sections, notice the use of the natbib \citep
129: %% and \citet commands to identify citations. The citations are
130: %% tied to the reference list via symbolic KEYs. The KEY corresponds
131: %% to the KEY in the \bibitem in the reference list below. We have
132: %% chosen the first three characters of the first author's name plus
133: %% the last two numeral of the year of publication as our KEY for
134: %% each reference.
135:
136: %\tableofcontents
137:
138: \section{Introduction}
139:
140: v892~Tau (Elias-1, Elias 3-1) is a young stellar object in the
141: Taurus-Aurigae star forming region \citep[$d=140~pc$,][]{kdh1994}.
142: Its visible spectrum is classified as spectral type B8V
143: \citep{hernandez2004},
144: %\footnote{Many papers in the literature report
145: % an errant spectral type of A6, which can be traced to a
146: % typographical error.},
147: making v892~Tau one of the few Herbig Ae/Be
148: stars in Taurus (in addition to AB~Aur, MWC 480). The spectral energy
149: distribution (SED) of v892~Tau is dominated by bright thermal emission
150: in the mid-infrared \citep{hillenbrand1992}, with relatively large
151: line-of-sight extinction in the visible \citep[one estimate is A$_{\rm
152: V}\sim5.9$,][]{kh1995}. Despite being one of the closest Herbig
153: stars and after decades of multi-wavelength observations, the nature
154: of v892~Tau is still uncertain.
155:
156: The SED suggests v892~Tau is either a young embedded Class I source
157: \citep{lada1987} still surrounded by its nascent envelope or a more
158: evolved, Class II object (i.e, Herbig Ae/Be star) seen through its
159: edge-on disk.
160: %Indeed, recent studies \citep{robitaille2006} suggest
161: %between 7-20\% of disk objects would appear embedded due to
162: %geometrical projection when based solely on SED
163: %analysis.
164: Spatially-resolved imaging can easily distinguish between
165: these two scenarios, and early speckle
166: interferometry \citep{kataza1991,haas1997,leinert2001} suggested the
167: presence of an extended and elongated nebula in the near-infrared that
168: could be due to scattering in bipolar lobes.
169:
170: The high-resolution speckle imaging of \citet{smith2005} clearly
171: resolved the K band ($\lambda=2.2\mu$m) emission to be coming from
172: two unresolved stars in v892~Tau with little or no sign of a nebula or
173: extended emission. The two stars of roughly equal brightness had an
174: apparent separation of 55~milliarcseconds (7.7~AU) and some evidence of orbital
175: motion was seen between epochs separated by 7 years.
176:
177: Although this binary should have a dramatic effect on the infrared
178: emission, carving out a large hole in the circumbinary disk,
179: recent analysis of the SED of v892~Tau including ISO data
180: \citep{acke2004} uncovered no distinct signature of the underlying
181: binary\footnote{\citet{acke2004} did note strong 11.2$\mu$m PAH emission and
182: anomalous infrared colors.}. Other workers \citep{liu2005,liu2007} made use
183: of the technique of nulling interferometry in the mid-IR
184: ($\lambda=10.3\mu$m) to marginally resolve v892~Tau (FWHM $\sim$20~AU)
185: along PA~164$\arcdeg$ consistent with normal (single-star) disk emission.
186:
187: Here, we report new mid-iIR imaging of the v892~Tau system which
188: resolves these mysteries, discovering very extended and resolved
189: emission that we interpret to be coming from a circumbinary disk.
190: We also confirm the presence of the binary at K
191: band and our new data allow first crude estimates of the orbital elements.
192: Lastly, we report a new SED analysis which
193: provides an improved determination of the system luminosity and our
194: viewing geometry.
195:
196:
197: \section{Observations}
198: \label{section:observations}
199:
200: We observed v892~Tau using the Keck-1 telescope as part of two
201: separate experiments, the Keck segment-tilting experiment and the Keck
202: aperture masking experiment. Here, we briefly describe these
203: experiments and give pertinent observing details.
204:
205: %\subsection{Keck segment-tilting experiment}
206: %
207: v892~Tau was observed on UT2004Aug31, UT 2004Sep01, and UT2005Feb19 at
208: the Keck-1 telescope using the 10.7$\mu$m filter
209: ($\lambda_0=10.7\mu$m, $\Delta\lambda=$1.55$\mu$m)
210: %\footnote{The PAH
211: %emission at 11.2$\mu$m reported by \citet{acke2004} is included
212: % within this filter's bandpass.}
213: of the Long Wavelength Spectrometer
214: \citep[LWS,][]{campbell2004} just before this instrument was
215: %regrettably
216: decommissioned in 2005. In order to optimize our
217: calibration against changes in seeing \citep[e.g.,][]{tuthill2000b},
218: we used the Keck's segmented primary mirror in a novel ``segment-tilting''
219: mode, whereby we controlled the tilts and pistons of the individual
220: mirror segments to create a set of 4 independent and non-redundant
221: interference patterns (using 6 segments each) on the LWS detector
222: focal plane. Furthermore, short exposures were used ($t_{\rm
223: int}=90$~ms$<t_0$, where
224: $t_0$ is the atmospheric coherence time $\sim$250~ms in the mid-iR) to effectively freeze the atmospheric turbulence.
225: For calibration we interleaved target observations with calibrators
226: $\alpha$~Tau\footnote{All calibrators were assumed unresolved except
227: $\alpha$~Tau where we adopted a uniform disk diameter of 19.8~mas
228: \citep{perrin1998}.} , $\iota$~Aur, and $\alpha$~Cet. Details on
229: this experiment and the implementation at Keck were first introduced
230: by \citet{monnier_aas2004} and more information can be found in
231: recently published science papers
232: \citep{weiner2006,ireland2007,rajagopal2007}.
233:
234: %See discussion of aperture masking by \citet{tuthill2000b} for a
235: %general overview of this technique and the practical advantages over
236: %conventional speckle imaging.
237:
238: %For this work, v892~Tau was observed on UT2004Aug31, UT2004Sep01, and
239: %UT2005Feb19 using th.
240: %Four patterns of 6~segments were tilted to give us non-redundant
241: %patterns of interference fringes allowing aperture synthesis imaging
242: %(note that in this method, light from 12 out of 36 segments is
243: %reflected away from the detector and is not used). Each
244: %pattern of 6~segments produces 15~interference fringes (baselines
245: %ranging between 2.6 and 9.3 meters) and 20~closure phases.
246:
247: The combined Fourier coverage of the four patterns over the three
248: nights (total of 6 independent pointings) can be found in
249: Figure~\ref{fig_vis2} along with the visibility results. The
250: visibility-squared data and the closure phases were compiled and saved
251: using the OIFITS data format \citep{pauls2005} and are available upon
252: request. Figure~\ref{fig_vis2} shows the
253: clear sign of highly-elongated disk emission.
254: % and the aperture
255: %synthesis imaging results will be discussed win
256: %\S\ref{section_imaging}.
257:
258: %\subsection{Keck aperture masking experiment}
259: %In addition to the mid-infrared observations described above,
260: We also
261: obtained diffraction-limited observations of v892~Tau in the
262: near-IR on UT2004Sep04 using the K-band filter
263: ($\lambda=$2.21$\mu$m, $\Delta\lambda=$0.43$\mu$m) of the NIRC camera
264: \citep{matthews96} in conjunction with an annulus aperture mask placed
265: in front of the secondary mirror. This observing mode was extensively
266: utilized between 1997 and 2005 \citep[see most recent science
267: papers,][]{monnier2007a, tuthill2008a} and details of
268: the experimental design and performance
269: %of the ``Keck aperture masking experiment''
270: can be found in \citet{tuthill2000b}. We interleaved
271: observations of the target with the unresolved calibrator 54~Per. The
272: Fourier coverage and visibility-squared results are shown in
273: Figure~\ref{fig_vis2}, where the the binary nature is clearly revealed.
274:
275: \section{Analysis}
276:
277: \subsection{Image Reconstructions}
278: \label{section_imaging}
279: We used the BSMEM image reconstruction software
280: \citep{buscher1994,lawson2004,lawson2006} for aperture synthesis
281: imaging. Figure~\ref{fig_images} shows the reconstructed images for
282: the mid-IR and near-IR data. In order to confirm the
283: asymmetric features in the mid-IR image were not due to artifacts in
284: the BSMEM algorithm, we also used the independent image reconstruction
285: code MACIM \citep{macim} and found good consistency between the basic
286: morphology, size scale, and level of emission asymmetry. As an
287: additional data quality check, separate images were made for each
288: independent night of data and the resulting images all closely
289: resembled the result shown in Figure~\ref{fig_images}. Note that we do not know
290: the relative position of the near-IR image compared to mid-IR image
291: and have presented each image centered in its corresponding
292: frame.
293:
294:
295: \subsubsection{Mid-infrared: Elongated disk structure}
296:
297: The mid-IR image shows elongated emission, approximately 320~mas
298: by 180~mas in full extent, elongated along PA~50$\arcdeg$. The two
299: bright lobes in the image are separated by 210~mas (30~AU) with the
300: south-west side being significantly brighter. A two-dimensional
301: Gaussian fit to the visibility data gives a FWHM 244$\pm$6~mas
302: $\times$ 123$\pm$9~mas along PA 49$\pm$1$\arcdeg$. For a simplistic
303: ``flat disk'' model, this 2-to-1 ratio elongation suggests
304: we are viewing this disk oriented at $\simge$60$\arcdeg$ inclination.
305:
306: A more suitable model for a circumbinary or transitional disk
307: would be a tilted and asymmetric ring of emission that
308: is nearly unresolved along the minor axis \citep[for example, see dramatic
309: case of HR 4796;][]{koerner1998,schneider1999}. Following the
310: parameterization of \citet{monnier2006}, we have fitted a ``skewed
311: asymmetric ring model'' to our interferometry data here to better estimate the
312: inner hole diameter.
313: The best fitting model had an inner-hole diameter of
314: 247$\times$121~mas (35$\times$17 AU) along PA 53$\arcdeg$, with a 40\%
315: skew along PA 284$\arcdeg$ (the thickness of the ring followed a
316: gaussian profile with a FWHM of 25\% ring radius). For purposes of
317: \S\ref{discussion}, we will assume the central hole in the
318: circumbinary disk has a radius of 17.5~AU.
319:
320: %We have confirmed the basic morphology using an independent image
321: %reconstruction code MACIM \citep{macim} although some differences were
322: %found. Specifically, the reconstructed image using MACIM had somewhat
323: %lower angular resolution and the emission was not clearly split into
324: %the two bright spots, but showed mainly a bright asymmetry to the
325: %south-west in more smooth, elongated emission. Otherwise, the size
326: %and orientation were compatible.
327:
328: Importantly, we want to emphasize that the scale of the emission is
329: more than 4$\times$ larger than the separation of the binary at the
330: heart of this system, and the elongation is along a
331: distinctly-different position angle. Since our mid-IR data in
332: 2004 were taken within one week of the near-IR data, we can
333: clearly prove that the elongated mid-IR emission comes from a
334: distinct and much larger component than the near-IR. We discuss
335: the relationship between the circumbinary disk and the underlying
336: binary orbit in \S\ref{discussion}.
337:
338: At first glance, it may seem surprising that \citet{liu2005} did not more
339: clearly resolve this large circumbinary disk using their nulling technique,
340: finding a FWHM of only 20~AU. However, this nulling measurement was
341: performed only for PA~164$\arcdeg$, which is along the narrow dimension
342: of the elongated emission we detected. Another possible explanation
343: for the smaller size is that the bandpass filter used here has more
344: contribution from 11.2$\mu$m PAH emission, which could be more extended
345: than the mid-IR continuum.
346:
347:
348: \subsubsection{Near-infrared: Binary star}
349: Figure~\ref{fig_images} shows the BSMEM image of v892~Tau at
350: 2.2$\mu$m. We confirm the binary nature of this target as first
351: reported by \citet{smith2005}. We also set a limit of $<$10\% of the
352: emission possibly coming from any sort of extended or ``halo''
353: component \citep[e.g.,][]{kataza1991,haas1997, leinert2001}. In order
354: to extract the separation and position angle of this binary, we have
355: used both image fitting and direct fitting to the interferometric
356: observables to yield the following result: $\rho=$44.2$\pm$1.0~mas,
357: $\theta=$79.9$\pm$1.0$\arcdeg$, Flux Ratio 1.15$\pm$0.04 (west
358: component brighter than east component).
359:
360:
361: \subsection{Binary orbit}
362: \label{orbit}
363: We combined our new binary measurement (at 2004.67) with the 1996.75 and 2003.76
364: measurements from \citet{smith2005} and these data are plotted in
365: Figure~\ref{fig_orbit}. The small proper motion ($\Delta\sim$10~mas$=1.4$~AU)
366: observed by \citet{smith2005} over 7~years is problematic for this
367: object since this would require a long period orbit ($\simge$100~yrs)
368: inconsistent with
369: the spectral types and luminosities of the binary stars themselves.
370:
371: %There are a few possible explanations for this anomalously
372: %small motion. Firstly, a very long orbital period naturally explains these
373: %observations but implies a system mass below 2$\msun$, which
374: %can be ruled out based on the known luminosity and stellar spectral types.
375: Alternatively, the binary period could be approximately 7~years, meaning the
376: stars had gone nearly exactly once around their orbit between epochs. Although
377: the timing is suspicious, the scenario is plausible except this
378: would require a system mass $>$20$\msun$, much too high for the B8
379: spectral type and system luminosity.
380:
381: The scenario we favor is that infrared variability affected the
382: brightness ratio between 1996 and 2003, confusing the assignment of
383: ``primary'' and ``secondary'' star. This variation could be intrinsic
384: or be due to varying line-of-sight obscuration through this asymmetric
385: disk. \citet{smith2005} did report a change in flux ratio, finding a
386: brightness ratio of 1.0 for 2003.76 which implies that their reported
387: position angle had a $\pm$180$\arcdeg$ ambiguity.
388:
389: In our 2004.67 data, we definitively detect the south-west component
390: as 15\% brighter, consistent with the PA assignment of the published
391: 2003.76 measurement. By flipping the position angle of the earlier
392: 1996 measurements, we find a robust family of orbits with periods of
393: $\sim$14~years, compatible with the measured binary separations and
394: the expected system mass. Note that future orbital refinement will
395: lead to precise stellar masses for these intermediate-mass,
396: pre-main-sequence stars \citep[see also MWC~361A,][]{monnier2006} -- a
397: rare opportunity to advance the calibration of pre-main-sequence
398: tracks for this mass range.
399:
400: To estimate the orbital parameters for v892~Tau, we used Monte Carlo sampling
401: of the measured stellar separations along with a system mass constraint of
402: 5.5$\pm$0.5$\msun$ expected for two stars of spectral type B8V \citep{palla1993}.
403: Figure~\ref{fig_orbit} shows the main results of our orbital study: $P=13.8\pm1.5$~yrs, $a=72.4\pm6.3$~mas (10~AU), $e=0.12\pm0.05$,
404: $i=60.6\pm3.8\arcdeg$, $\omega=233\pm42\arcdeg$, $\Omega=28\pm5\arcdeg$,
405: $T_O = 55480\pm900$~MJD.
406:
407: \section{Discussion}
408: \label{discussion}
409: %% mention the strong PAH emission (acke2004).
410: We interpret the mid-IR emission of v892~Tau as a circumbinary
411: disk based on several arguments. The mid-IR emitting region is much
412: larger than the separation between the binary components, ruling out
413: models which have all the mid-IR emission coming from disks around the
414: individual stars. In theoretical models of tidal truncation
415: \citep{artymowicz1994}, the circumbinary disk has an inner hole
416: approximately 1.8-2.6$\times$ larger than the semi-major axis of the
417: binary system (for eccentricity between 0 and 0.25). For comparison,
418: our best estimates of the hole radius and orbital semi-major axis are
419: 17.5 AU and 10~AU respectively -- giving a ratio of 1.75 which is
420: close to the theoretical expectation. Additionally, the position
421: angle and derived inclination angle of the mid-IR emission
422: (PA$\sim$50$\arcdeg$, $i\sim$60$\arcdeg$)
423: is similar to
424: those derived from the orbit (PA$\sim$28$\arcdeg$, $i\sim$61$\arcdeg$)
425: and the mild
426: asymmetry in the mid-IR emission suggests dynamical interactions
427: between an eccentric binary and the surrounding disk (via resonances or disk warping).
428:
429:
430: We further test the circumbinary disk hypothesis by analyzing the
431: spectral energy distribution (SED). The SED shows a large mid-IR bump
432: similar to that seen in ``transitional'' disks
433: \citep{calvet2002,espaillat2007}. By reddening template Kurucz spectra for
434: B8V stars, we can fit the near-IR emission with mostly photospheric
435: light (some near-IR emission from hot dust is allowed but not well
436: constrained) assuming A$_V$ = 10.95 yielding a combined stellar
437: luminosity of 400~L$_{\odot}$, reasonable for two B8V stars. Our new
438: proposed SED decomposition for v892~Tau can be found in
439: Figure~\ref{fig_orbit}, showing that
440: the rising mid-IR flux can be roughly characterized as a
441: T$\sim450$~K blackbody, consistent with emission from
442: the warm inner wall of the circumbinary disk.
443:
444: Using equation (1) in \citet{isella2006}, the predicted radius for the
445: warm inner wall for the above stellar luminosity (assuming 0.25 $\mu$m spherical
446: silicate grains) is 18~AU -- consistent with our derived inner hole radius
447: of 17.5~AU based on imaging. Furthermore, we find that the scale
448: height of this wall at 18 AU is $\sim$ 1.8 AU \citep{ddn2001} which
449: can produce the observed line-of-sight A$_V \sim$ 11 for the disk
450: inclination of $\sim$65$^o$. This latter phenomena may not be widely
451: appreciated -- circumbinary (or transitional) disks have warm
452: puffed-up inner walls (the rim scale height is a stronger function of
453: radius than temperature) at large radii that effectively increases the
454: possibility that central stars will be obscured or at least reddened
455: as seen by distant observers. The lack of scattering nebulosity in archival Hubble Space Telescope images
456: independently suggests that the high line-of-sight A$_V$ is likely from absorption by dust in the outer disk and not infalling envelope material.
457: While a detailed model is beyond the
458: scope of this Letter, a careful study will allow precise
459: constraints of stellar luminosities, dust properties, and circumbinary
460: structure and should be straight-forward with today's 3-D Monte Carlo
461: radiative transfer codes \citep[e.g., TORUS,][]{torus}.
462:
463: \section{Conclusions}
464: We have discovered an extensive circumbinary disk around v892~Tau in
465: the mid-infrared. We also independently confirm the binary nature of
466: the underlying stellar system and our new measurement allows us to fit
467: an astrometric orbit, finding a period $\sim$14~years for system mass of $\sim$6$\msun$.
468: Our limited orbital phase
469: coverage and some ambiguity in position angles allow only
470: crude estimates and we strongly urge continued monitoring of this
471: system using near-IR speckle, aperture masking, or adaptive optics.
472:
473: We have proposed a new SED decomposition with a line-of-sight
474: extinction ($A_V\sim11$) higher than previously thought, implying a
475: system luminosity $\sim$400$\lsun$. This result highlights
476: the fact that circumbinary disks (and transitional disks) have much
477: larger opening angles since the puffed-up inner wall causes
478: enhanced extinction of the central stars even at
479: inclinations of 60$\arcdeg$.
480:
481: v892~Tau is another case where high-resolution imaging has motivated a
482: fundamental shift in our understanding of an individual object. As
483: spectral energy distributions are used to discover ``transitional''
484: disks implicating planet formation, we must be cognizant of the
485: important role of binarity and the associated circumbinary disks that
486: can mimic signs of planet formation \citep[e.g.,][]{ireland2008}.
487: Far from merely being spoilers to planet finders, new circumbinary
488: disks offer fresh laboratories for studying dynamical interactions
489: between gas-rich disks and the massive orbiting bodies embedded within
490: them as well as critical opportunities to calibrate pre-main sequence
491: tracks.
492:
493: \acknowledgments {We thank Nuria Calvet and Jesus Hernandez for
494: helpful discussions and acknowledge support from NASA KPDA grant
495: 1267021, NASA Origins grant NNG05GI80G, NSF-AST 0352728, and the
496: NASA Michelson Fellowship program (MJI). The data presented herein
497: were obtained at the W.M. Keck Observatory, which is operated by
498: Caltech, University of California and NASA. WMKO
499: was made possible by the financial support of the W.M. Keck
500: Foundation. The authors wish to recognize and acknowledge the very
501: significant cultural role and reverence that the summit of Mauna Kea
502: has always had within the indigenous Hawaiian community. We are
503: most fortunate to have the opportunity to conduct observations from
504: this mountain.
505:
506: {\it Facility:} \facility{Keck:I (LWS,NIRC)} \facility{Hiltner (TIFKAM)}
507: }
508:
509: \bibliographystyle{apj}
510: \bibliography{apj-jour,ms}
511: %v892tau,RS_Oph,Review2,Review,IONIC3,iKeck,KeckIOTA}
512:
513: \clearpage
514: \begin{figure}[hbt]
515: \begin{center}
516: \includegraphics[angle=90,width=7in]{f1.eps}
517: \figcaption{\footnotesize This figure shows visibility
518: data (color scale) for v892~Tau and the specific Fourier coverage (dots).
519: The left panel shows results for 10.7$\mu$m using the Keck segment tilting method while the right panel applies to the 2.2$\mu$m Keck aperture masking data.
520: \label{fig_vis2}}
521: \end{center}
522: \end{figure}
523:
524: \clearpage
525: \begin{figure}[hbt]
526: \begin{center}
527: \includegraphics[angle=90,width=7in]{f2.eps}
528: \figcaption{\footnotesize This figure shows the image reconstructions
529: of v892~Tau at 10.7$\mu$m (left panel) and 2.2$\mu$m (right panel)
530: using the BSMEM software. At the bottom-left of each panel, we have
531: included an estimate of the resolution of each image, corresponding
532: here to 80~mas at 10.7$\mu$m and 16~mas at 2.2$\mu$m. Contour
533: levels are shown for the extended emission in the mid-IR
534: spaced logarythmically in factors of 4: 25\%, 6.3\%, and 1.6\% of
535: the peak. The scale bar applies for a distance of 140~pc.
536: For orientation, East points left and North points up.
537: \label{fig_images}}
538: \end{center}
539: \end{figure}
540:
541: %\clearpage
542: % \begin{figure}[hbt]
543: % \begin{center}
544: % \includegraphics[angle=90,width=7in]{Figures/fig3.epsi}
545: % \figcaption{\footnotesize {\em (left panel)} This figure shows the
546: % a simple skewed ring model for the mid-infrared emission
547: % of v892~Tau. This model suggests the circumbinary disk has an inner hole with
548: % diameter 250~mas (35~AU). {\em (right panel)} The original ring model has been
549: % convolved to an angular resolution of 80~mas in order to better compare
550: % the model features with the observed images.
551: % \label{fig_ring}}
552: % \end{center}
553: % \end{figure}
554:
555: \clearpage
556: \begin{figure}[hbt]
557: \begin{center}
558: \includegraphics[angle=90,height=2.5in]{f3a.eps}
559: \includegraphics[angle=90,height=2.5in]{f3b.eps}
560: \figcaption{\footnotesize {\em (left panel)} This figure shows the
561: separation and position angle measurements for v892~Tau from
562: \citet{smith2005} and this work. We have plotted the best-fit model
563: for a system mass of 5.5~$\msun$ (period 13.8 yrs, eccentricity 0.12) with the solid curve.
564: Note that we had to flip the position angle of the 1996 epoch
565: measurement in order to find a realistic orbit (see \S\ref{orbit} for full description of our
566: procedure and justification). {\em (right panel)} This figure shows
567: the spectral energy distribution for v892~Tau
568: \citep[photometry from][and our own JHK photomery from MDM Observatory]{strom1994,acke2004,herbst1999}
569: and our proposed
570: decomposition of star plus disk. We have plotted the contribution
571: from the reddened ($A_V=10.95, R_V=6.4$) stellar photospheres
572: (2$\times$B8V, system luminosity of 400$\lsun$) and the contribution
573: from the ``warm inner wall'' of the circumbinary disk with
574: temperature T$\sim$450K.
575: \label{fig_orbit}}
576: \end{center}
577: \end{figure}
578:
579:
580:
581: \end{document}
582: