Free Access
Issue
A&A
Volume 578, June 2015
Article Number A1
Number of page(s) 7
Section Stellar structure and evolution
DOI https://doi.org/10.1051/0004-6361/201525634
Published online 22 May 2015

© ESO, 2015

1. Introduction

Brown dwarfs (BDs) are the bridge between late-type stars and planetary-mass objects. T dwarfs (with effective temperatures of Teff ~ 1500–500 K) and Y dwarfs (Teff ≤ 500 K ) are among the coolest and least luminous objects detected, and their properties can help us to understand giant extrasolar planets.

Table 1

Sample of observed brown dwarfs.

In recent years, a significant number of BDs have been identified in wide-field optical and infrared imaging surveys, e.g., the Two Micron All Sky Survey (2MASS, Skrutskie et al. 2006), the DEep Near Infrared Survey (DENIS, Epchtein et al. 1997), or the UKIRT Infrared Deep Sky Survey (UKIDSS, Lawrence et al. 2007). For the coolest objects, deep near-IR and mid-IR surveys such as the Canada-France Brown Dwarf Survey (CFBDS, Delorme et al. 2008) or the Wide-Field Infrared Survey Explorer (WISE, Wright et al. 2010) have contributed to significantly increasing the number of known T and Y dwarfs (e.g., Burningham et al. 2010; Albert et al. 2011; Kirkpatrick et al. 2011, 2012; Cushing et al. 2011; Tinney et al. 2012; Pinfield et al. 2014a,b). Currently, there are ~355 and 15 identified T and Y dwarfs, respectively1.

Multiplicity is directly related to the stellar and substellar formation process, and it has been studied for different mass regimes. In the case of T dwarfs, several works have focused on studying the companions around these sources (e.g., Burgasser et al. 2003, 2006; Bouy et al. 2003, 2008; Gelino et al. 2011; Liu et al. 2011, 2012). Most of them have benefited from laser guide star (LGS) adaptive optics (AO) assisted observations at large ground-based telescopes or/and the Hubble Space Telescope (HST) data in order to achieve the best angular resolution and sensitivity. As a result there are now ~14 identified T-dwarf binaries. Most of them show very small separations (~2–5 au, Burgasser et al. 2003, 2006; Liu et al. 2011), although there are also wide binaries that have been detected (e.g., WISEJ1711+3500, ρ = 15 ± 2 au, Liu et al. 2012). In a recent work, Aberasturi et al. (2014) have derived a binary fraction of <16 or <25%, depending on the underlying mass ratio distribution.

Motivated by the large number of T and Y dwarfs identified by WISE, we started a program to search for companions to newly discovered BDs in 2012 using high angular resolution LGS AO-assisted observations in the near-IR. Most of the selected targets were T and Y dwarfs identified with WISE and listed in Kirkpatrick et al. (2011). In this paper we present the results from our AO imaging campaigns. In Sects. 2 and 3 we describe the sample selection, the observations, and the data reduction. In Sect. 4 we discuss the main results, including the sensitivity limits. The main conclusions are summarized in Sect. 5.

Table 2

Observing log.

2. The sample of brown dwarfs

Our target list was mainly selected from the newly found BDs with WISE, which is a mid-IR satellite launched in 2009 that surveyed the entire sky in four bands: 3.4, 4.6, 12, and 22 μm. Finding BDs was a primary science goal of WISE, and the strategy was based on a particular color cut: the methane absorption feature near 3.3 μm made the [3.6][4.6] color extremely red (2 mag), unique among substellar objects.

Kirkpatrick et al. (2011) summarize the first WISE BDs discoveries that included a list of a hundred of detected substellar objects. For our study, we initially selected WISE T and Y dwarfs from this work. We kept only those observable from Paranal Observatory and never studied for multiplicity before (Gelino et al. 2011; Liu et al. 2012).

Additional dwarfs were discovered and studied afterwards (Cushing et al. 2011; Gizis et al. 2011; Kirkpatrick et al. 2012), and we included them in our target list. Finally, we also considered several bright targets (mainly previously known L-type and early-T dwarfs) as backups to be observed under bad conditions. One of these targets, 2MASS J1114-2618, was observed by the HST in the meantime, and the results were recently presented by Aberasturi et al. (2014). We include it here for comparison. In total, we selected ~50 targets for our study.

3. Observations and data reduction

The observations presented here were obtained with NAOS-CONICA (NACO), the AO system and near-IR camera (Lenzen et al. 2003; Rousset et al. 2003) at the Very Large Telescope (VLT). Since most of the targets are too faint to be used as natural guide stars (NGS), we made use of the laser guide star (LGS) facility at the VLT.

To obtain a full AO correction, the LGS facility should be used in combination with a tip-tilt (tt) reference star brighter than V = 17 mag, located at a maximum separation of 40 from the target. In the absence of a tt star, the LGS can be used in the so-called “seeing enhancer” (se) mode (Girard et al. 2010), that is, LGS-assisted AO observation but without correcting for tip/tilt. Because the low orders of the wavefront corrugations are not corrected, the SE mode provides an image quality intermediate between simple (non-AO) imaging and full LGS closed loop observations. (We refer to the NACO manual for a full description of this mode2.) For our observations, we used a tt reference star whenever available, and switched to se mode otherwise.

thumbnail Fig. 1

Sensitivity limit of the NACO observations. We show the curves for the five apparently single sources, one target per row, in the different NACO filters. They were computed as the 3σ standard deviation of the PSF radial profile.

Open with DEXTER

The observations were allocated in seven different nights over 2012 and 2013. We could not collect any data in four of the nights (2012-06-10, 2012-07-12, 2012-12-08, and 2013-03-07) because of inadequate atmospheric conditions for AO and the LGS and/or technical issues associated to the laser itself. The atmospheric conditions in the three other nights were moderate to poor for the LGS, with average coherence times (t0) ≤ 3 ms and variable seeing. In these conditions we observed a total of 20 targets, but we could only obtain good quality datasets for seven of them. They are included in Table 1. As shown in the table, some of the targets were even observed without AO owing to the unsuitable atmospheric conditions and poor LGS performance.

All the targets were observed with the S54 objective, providing a field of view of 54″ × 54″ and a nominal plate scale of 0.̋054. The data was obtained in one or several near-IR filters (JHKs), depending on the atmospheric conditions and brightness. To estimate and remove the sky background, we observed the targets applying random offsets between the individual exposures (“Autojitter” mode).

The data was reduced using eclipse (Devillard 1997) and following the standard steps for near-IR imaging data: the images were dark-subtracted and flat-field-divided, and bad pixels were masked out. The sky background was estimated and subtracted from all the individual images. Finally, the frames were shifted and co-added to generate the final stacked image. Since the atmospheric conditions were variable during the observations, some of the individual frames showed a low image quality. To improve the final spatial resolution of the combined data, we removed the worst individual images, i.e. those with the lowest Strehl ratio and the largest FWHM, and kept only those with the best AO performance.

The observing log is given in Table 2. As can be seen, the seeing and coherence time were not optimal for AO+LGS observations, and this is reflected in the final FWHM and Strehl ratios measured on the combined images.

4. Results

Five of the seven BDs observed with NACO+LGS are apparently single. We estimated the sensitivity limit as a function of the separation for all of them (see Fig. 1) in the different NACO filters. The curves have been estimated from the point spread function (PSF) radial profiles of the targets, while computing the standard deviation of the noise in bins of 20 pixels and one pixel step. We derived the difference in magnitudes at a 3σ level normalizing by the peak intensity of the PSF.

Usually, the best combination of sensitivity and angular resolution is achieved in the H band. To derive the sensitivity limits in this band, we used the H mag listed in Table 1. For the targets with MKO filter system photometry, we converted the magnitudes into the 2MASS system using the relations from Stephens & Leggett (2004). We obtain H2MASS ~ HMKO−0.04 mag for the given spectral types. We reach 3σ magnitude limits between H~19–21 mag for separations of 0.̋5 (see Table 3). For an average distance of ~15 pc and ages of 1 Gyr and 5 Gyr, this magnitude interval corresponds to masses of ~117 MJup, and ~30–18 MJup, respectively, according to COND evolutionary tracks (Chabrier et al. 2000). For each individual target, we estimated the approximate spectral type (Sp. Type lim) that corresponds to the H-band limiting magnitude (3σ) using the MH versus spectral type relation derived by Kirkpatrick et al. (2012), and the individual distances from Table 1. This value is included in the last column of Table 3. In the case of WISE 1114-2618, we achieve a ΔH value of 4.7 mag at 0.̋5. This is deeper than the HST/F164N observations presented by Aberasturi et al. (2014), with a contrast of 1.5 mag at a separation of 0.̋6.

Table 3

Contrast and sensitivity limits (3σ) in the H-band at a separation of 0.̋5.

thumbnail Fig. 2

NACO/LGS observations of WISE 0612-3036. The top panel shows the ~ 54″ × 54″ NACO FOV. The target, WISE-0612-3036, is in the center of the black square. The star 2MASSJ06121227-3036100, used to calibrate the JHKs photometry of the binary system, is the brightest source in the west. The bottom panels show a zoom on the newly discovered binary in JHKs.

Open with DEXTER

4.1. WISE 0612-3036

We have resolved WISE 0612-3036 (WISE0612, hereafter) into a subarcsecond binary (Fig. 2). The two objects are detected in the three bands, JHKs with an estimated separation of ρ = 350 ± 5 mas and position angle of PA = 235 ± 1°. The values are derived computing the average and standard deviation of the measurements in the three bands. We note that only the H-band data was obtained with LGS assisted AO, while the subsequent JKs data was obtained in open loop owing to technical problems with the laser.

We performed PSF stellar photometry of the binary system with ALLSTAR in DAOPHOT II (Stetson 1992). For the photometry we used the final stacked images in each filter. The PSF was constructed using three or four stars in the field of view for the J and HKs bands, respectively. The typical PSF errors of both components are in the range of 0.012–0.034 mag. For the transformation to the standard 2MASS JHKs system, we used the only star in the field with 2MASS photometry: 2MASS J06121227-3036100 with and and apparent magnitudes of J = 15.915 ± 0.080 mag, H = 15.141 ± 0.066 mag, and Ks = 15.024 ± 0.139 mag. The final errors of the photometry are weighted means of the PSF errors and errors of the standard star. To double check the PSF photometric calibration with the bright star 2MASS J06121227-3036100, we estimated its Ks-band magnitude using a standard star observed at the end of the night. We find that the difference between the 2MASS tabulated value and our calibrated photometry is only 0.04 mag.

thumbnail Fig. 3

Color–color near-IR diagram. We have included a sample of ultracool dwarfs from Dupuy & Liu (2012): the small gray diamonds represent M and L dwarfs, while the green diamonds are T dwarfs. We have only plotted objects with photometric uncertainties smaller or equal to 0.2 mag. The binary components of WISE0612 are represented by blue triangles.

Open with DEXTER

The estimated magnitudes for the two components in the 2MASS system are provided in Table 4. Kirkpatrick et al. (2011) report magnitudes of J = 17.00 ± 0.09, H = 17.06 ± 0.11, and Ks = 17.34 ± 0.21 for the unresolved system. If we correct them for binarity, the difference between our PSF photometry and these values are −0.12, 0.14, and 0.16 in JHKs, respectively, which are consistent within the given uncertainties.

The primary (A) is the brightest object located to the NE (see Fig. 2). Kirkpatrick et al. (2011) classify the unresolved object as a T6 dwarf based on near-IR spectroscopy. The difference in magnitudes between the two components are ΔJ = −0.05 ± 0.14 mag, ΔH = −0.10 ± 0.14 mag, and ΔKs = −0.13 ± 0.24 mag, so we can assume similar spectral types for the two objects.

We have plotted the binary components in a near-IR colorcolor diagram (see Fig. 3) and compared them with well known T-type dwarfs from Dupuy & Liu (2012, Table 10) with available 2MASS photometry. We have only included objects with photometric uncertainties smaller than 0.2 mag in the three bands. The WISE0612 system falls within the region of the late-T dwarfs with a very blue HKs color. As explained in Burgasser et al. (2002), this is expected in late-T dwarfs since the 2.1 μm peak starts to flatten until it is suppressed for the latest type objects. This is indeed the case of WISE0612, which shows no emission peak at 2.1 μm in the near-IR spectrum obtained by Kirkpatrick et al. (2011).

Table 4

Properties of the binary system WISE 0612-3036 AB.

The spectrophotometric distance to the unresolved source, based on its WISE W2 magnitude and its spectral type, is ~17 pc (Kirkpatrick et al. 2011). To derive the distance to the new binary system we have used the mean 2MASS absolute magnitudes as a function of the spectral type from Dupuy & Liu (2012). The values derived for a T6 dwarf are MJ = 15.35 ± 0.02 mag, MH = 15.32 ± 0.02 mag, and MKs = 15.68 ± 0.01 mag, with typical dispersions of 0.11, 0.17, and 0.33 mag, respectively. Using these values, we derived distances of ~31 ± 2 pc (J-band), 29 ± 3 pc (H-band) and 32±5 pc (Ks-band) and adopted an average value of d = 31 ± 6 pc for the rest of the paper.

Assuming that the two objects are bound, we represent the binary components in several color-absolute magnitude diagrams (see Fig. 4), together with a sample of MLT dwarfs with the measured parallaxes included in Dupuy & Liu (2012). We excluded objects with 2MASS photometric errors larger than 0.5 mag, and relative errors in their parallaxes (δπ/π) larger than 0.6. As expected, the two binary components of WISE0612 lie in the region between T5-T7 dwarfs, with very blue HKs and JKs colors.

For the assumed distance of 31 ± 6 pc, the projected separation of the binary is ρ ~ 11 ± 2 au. Since the age of the system is unknown, we estimated the mass of the objects assuming two ages of ~1 Gyr and 5 Gyr. According to COND evolutionary tracks (2MASS system, Chabrier et al. 2000), we derived masses of ~30 MJup (1 Gyr) and ~60 MJup (5 Gyr) for the individual components. For a circular orbit, we derived orbital periods of ~150 yr for 1 Gyr and ~105 yr for 5 Gyr.

Finally, we note that a single-epoch observation is not enough to classify the binary as bound. Although additional data is needed to confirm it as a common proper motion pair, the small angular separation makes a random coincidence unlikely. We have estimated the probability of a background object lying close to the target using the number sources detected in the NACO H-band image. We detect eight sources in the ~56′′×56′′ field of view. Since the companion candidate is detected at a separation of 0.̋35, the probability of detecting an object in an area of arcsec2 around the target is 8 × (0.3848/562), that is, 0.09%. If we only consider the detected sources with a similar brightness to the companion (2), we derive a probability of 0.02%. We have also checked that there is no object at the given position in the Palomar Observatory Sky Survey (POSS) II images.

On the other hand, the proper motion of the primary estimated by Kirkpatrick et al. (2011) is μα.cosδ = −137 ± 22 mas/yr and μδ = −256 ± 21 mas/yr, that is, ~0.3′′/yr. Therefore, the binary can be confirmed with second-epoch non-AO observations obtained after 34 yrs, to allow the BD enough time to move far enough from the companion, if it turns out that the latter is not co-moving.

thumbnail Fig. 4

Near-IR color–magnitude diagrams. We represent a sample of ultracool dwarfs with measured parallaxes and 2MASS photometry from Dupuy & Liu (2012): gray diamonds represent ML-dwarfs and green diamonds the T dwarfs. We have highlighted in red the T-dwarfs with spectral types between T5 and T7. The binary components of WISE0612 are represented by blue open triangles. In the lower central panel, the faint companion candidate to WISE2255 is represented by a filled square and WISE 1711+3500B (T9.5, Liu et al. 2012) by a filled circle. A sample of 4 Y dwarfs (Liu et al. 2012; Leggett et al. 2013) are indicated by asterisks.

Open with DEXTER

4.2. WISE 2255-3118

In the case of the T8 dwarf WISE J2255-3118 (WISE2255, hereafter), the final image quality was not good enough to detect very close companions. However, we have detected a very faint source in the Ks band at a separation of ~1.̋3 (see Fig. 5). This source is marginally detected in the H band. The estimated distance to the source, 13 pc (Kirkpatrick et al. 2011), implies a projected separation of ~17 au.

We performed aperture photometry on the HKs-band images and calibrated it using standard stars observed during the night. We derive H = 17.64 ± 0.18 mag and Ks = 17.47 ± 0.25 mag for the primary. For the faint object we estimate Ks = 20.5 ± 0.3 mag and H> 20.6 mag and a color of HKs > 0.1 mag.

If bound, the companion candidate would be at a distance of 13 pc, which implies a MKs ~ 19.9 mag. When we compare its properties to those of the T dwarfs represented in the central lower panel of Fig. 4, we see that the object does not follow the sequence of the coldest T dwarfs, displaying a redder HKs color. In this panel we also compared WISE2255 with WISE 1711+3500, a physical binary system with very similar properties: it consists of a T8 primary and a T9.5 companion with ΔK = 3 mag (Liu et al. 2012). The difference in the H band between the two components is ΔH of −2.83, and its HKs,MKO color is −0.42 ± 0.17. Using the relations by Stephens & Leggett (2004) for a T9 dwarf (the coldest object in their sample), we can convert its magnitudes into the 2MASS system (Ks,2MASS = Ks,MKO + 0.16, and H2MASS = HMKO + 0.04), obtaining HKs,2MASS = − 0.54, significantly bluer than the upper limit reported for the WISE2255 companion candidate.

thumbnail Fig. 5

NACO/LGS observations of WISE 2255-3118. We display a field of 8″ × 8′′. A faint source (encircled) is detected at ~1.̋3 SE from the T dwarf.

Open with DEXTER

Finally, we also compared the photometry of the companion candidate with five Y dwarfs included in Liu et al. (2012) and Leggett et al. (2013). Since they all have HKs photometry in the MKO system, we have converted them to the 2MASS system following the same procedure described above. The result is displayed in Fig. 4 (low central panel). The companion is redder than the Y dwarfs, but it lies close enough to their locus, preventing us from drawing a firm conclusion about the nature of this object.

We searched in public archives for additional data on this source and found HST/WFC3 data obtained in two near-IR filters, F110W and F160W (Program 12972, P.I. Gelino). The observations were obtained on 2012 October 17 in MULTIACCUM mode and following a four dithering pattern. The total exposure time was ~1200 s in the two filters. We retrieved the final calibrated images (flt name extension) from the Hubble Legacy Archive (HLA) and reprocessed them using the routines within the DrizzlePac3 package. In particular, we used multidrizzle to combine the exposures into a master image, resampling onto a finer pixel scale and correcting for geometric distortions in the camera. (We selected a final_scale of 0.09′′/pixel and final_pixfrac of 0.8.) The T dwarf and the companion are clearly detected in the final HST images (see Fig. 6), and the analysis of the two images with SEXtractor (Bertin & Arnouts 1996) and PSFEx (Bertin 2011) reveals that the faint companion candidate is an extended source. To visualize it, we represented the HST PSF-subtracted images in Fig. 6, where the companion shows strong residuals in comparison to the T dwarf. This is consistent with the concentration index (CI) derived by the HST team and included in the HLA SEXtractor catalog for the F160W band (Data Release 8). This index is defined as the difference in magnitude from two different apertures (0.̋09 and 0.̋27), and it is used to separate extended from point-like sources. They measure a CI =1.4 for the companion candidate, while point-like sources in the WFC3 show 0.5 < CI < 1.0. We note that the HST image is relatively crowded, with a large number of spatially resolved galaxies. In fact, the HLA catalog contains 275 detected sources, with 235 displaying CI > 1.0.

An extended source is consistent with a background galaxy or with a binary. In the latter case, and if bound to the primary, the binary components would be cooler than T8, and probably in the Y dwarf regime. However, with the current data we cannot distinguish between the two scenarios.

That the detected source is extended, together with the red HKs color, suggests that the faint object is most probably a background galaxy, although additional imaging is needed to test whether the objects are co-moving or not.

thumbnail Fig. 6

HST/WFC3 F110W and F160W reduced images of WISE 2255-3118. We display a field of 8′′×8′′. North is up and east to the left. The companion candidate is clearly resolved in the two filters. The PSF-subtracted images are displayed on the right panels, where the companion candidate shows large residuals consistent with an extended source.

Open with DEXTER

Table 5

Summary of known T-dwarf binary systems.

thumbnail Fig. 7

Known T-dwarf binaries. We have represented the spectral types of the secondaries versus the primaries for known T-dwarf binaries (circles). 0 stands for T0, 10 for T10, and 11 for Y0. The color code is related with the projected separation (in au) of the binaries. Most of them show separations smaller than 5 au, while only four show ρ ≥ 8 au. WISE 0612-3036 (filled square) is among these four binaries with a projected separation of 11 au.

Open with DEXTER

4.3. Comparison with T-dwarf binaries

To put our study into context, we compared the properties of WISE 0612-3036 with other T-dwarf binaries from different AO and HST studies (see Table 5). We have represented the spectral types of the secondaries versus the spectral types of the primaries in Fig. 7. Their projected separation in the sky (in au) have been color-coded. As seen in Fig. 7, all the early-T primaries with SpT < T3 have secondaries later or equal than T4, while most of the late-T primaries (SpT T4) show nearly equal mass companions (typical errors in spectral typing are between 0.51 type). The separations vary between ~116 au, with only four binaries showing values larger than or equal to 8 au. WISE 0612-3036 is among these four binaries with a projected separation of 11 au.

5. Conclusions

We have observed seven BDs with the NACO+LGS system to look for close companions. In general, the observing conditions were not optimal for the LGS, resulting in poor corrections that did not allow the innermost regions around the targets to be explored.

Our main findings can be summarized as follows:

  • Five targets are apparently single with 3σ limiting magnitudes of H between 19–21 mag at 0.̋5.

  • The T-dwarf WISE 0612-3036 is clearly resolved into a subarsecond binary. PSF-photometry of the binary has allowed us to derive the individual JHKs magnitudes of the two components, which are consistent with a very similar spectral type of T6. Using absolute magnitude-spectral type relationships in JHKs, we estimate a distance to the binary of d ~ 31 ± 6 pc, which implies a projected separation of ρ = 11 ± 2 au. Additional observations are needed to confirm the system as bound.

  • The target WISE 2255-3031 shows a very faint source at a separation of 1.̋3 SE from the primary. Its HKs color (>0.1 mag) is redder than the colors of the latest T dwarfs and Y dwarfs. Since HST public archival data in two near-IR filters shows that the source is extended, we suggest that it is most probably a background galaxy.

  • Similar to other known late T-dwarf binaries, WISE 0612-3036 is comprised of two nearly equal-mass objects. Its projected separation, 11 au, is larger than the average value for T dwarf binaries (~5 au).


Acknowledgments

We thank the referee, C. Reyle, for her useful comments. This research has been funded by Spanish grants AYA2010-21161-C02-02 and AYA2012-38897-C02-01. Support for N.H., J.B., R.K., and D.M. is provided by the Ministry of Economy, Development, and Tourism’s Millennium Science Initiative through grant IC120009, awarded to The Millennium Institute of Astrophysics, MAS. J.B. is supported by FONDECYT No.1120601, R.K. is supported by Fondecyt Reg. No. 1130140. D.M. is also supported by FONDECYT No.1130196 and by the Center for Astrophysics and Associated Technologies CATA PFB-06. C.C. acknowledges the support from project CONICYT FONDECYT Postdoctorado 3140592. We are indebted to the Paranal staff for their support during the runs. N.H. thanks H. Bouy for useful discussions. This research has benefited from the M, L, T, and Y dwarf compendium housed at DwarfArchives.org.

References

All Tables

Table 1

Sample of observed brown dwarfs.

Table 2

Observing log.

Table 3

Contrast and sensitivity limits (3σ) in the H-band at a separation of 0.̋5.

Table 4

Properties of the binary system WISE 0612-3036 AB.

Table 5

Summary of known T-dwarf binary systems.

All Figures

thumbnail Fig. 1

Sensitivity limit of the NACO observations. We show the curves for the five apparently single sources, one target per row, in the different NACO filters. They were computed as the 3σ standard deviation of the PSF radial profile.

Open with DEXTER
In the text
thumbnail Fig. 2

NACO/LGS observations of WISE 0612-3036. The top panel shows the ~ 54″ × 54″ NACO FOV. The target, WISE-0612-3036, is in the center of the black square. The star 2MASSJ06121227-3036100, used to calibrate the JHKs photometry of the binary system, is the brightest source in the west. The bottom panels show a zoom on the newly discovered binary in JHKs.

Open with DEXTER
In the text
thumbnail Fig. 3

Color–color near-IR diagram. We have included a sample of ultracool dwarfs from Dupuy & Liu (2012): the small gray diamonds represent M and L dwarfs, while the green diamonds are T dwarfs. We have only plotted objects with photometric uncertainties smaller or equal to 0.2 mag. The binary components of WISE0612 are represented by blue triangles.

Open with DEXTER
In the text
thumbnail Fig. 4

Near-IR color–magnitude diagrams. We represent a sample of ultracool dwarfs with measured parallaxes and 2MASS photometry from Dupuy & Liu (2012): gray diamonds represent ML-dwarfs and green diamonds the T dwarfs. We have highlighted in red the T-dwarfs with spectral types between T5 and T7. The binary components of WISE0612 are represented by blue open triangles. In the lower central panel, the faint companion candidate to WISE2255 is represented by a filled square and WISE 1711+3500B (T9.5, Liu et al. 2012) by a filled circle. A sample of 4 Y dwarfs (Liu et al. 2012; Leggett et al. 2013) are indicated by asterisks.

Open with DEXTER
In the text
thumbnail Fig. 5

NACO/LGS observations of WISE 2255-3118. We display a field of 8″ × 8′′. A faint source (encircled) is detected at ~1.̋3 SE from the T dwarf.

Open with DEXTER
In the text
thumbnail Fig. 6

HST/WFC3 F110W and F160W reduced images of WISE 2255-3118. We display a field of 8′′×8′′. North is up and east to the left. The companion candidate is clearly resolved in the two filters. The PSF-subtracted images are displayed on the right panels, where the companion candidate shows large residuals consistent with an extended source.

Open with DEXTER
In the text
thumbnail Fig. 7

Known T-dwarf binaries. We have represented the spectral types of the secondaries versus the primaries for known T-dwarf binaries (circles). 0 stands for T0, 10 for T10, and 11 for Y0. The color code is related with the projected separation (in au) of the binaries. Most of them show separations smaller than 5 au, while only four show ρ ≥ 8 au. WISE 0612-3036 (filled square) is among these four binaries with a projected separation of 11 au.

Open with DEXTER
In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.