Near-Infrared Spectroscopy of Very Low-Luminosity Young Stellar Objects in the Taurus Molecular Cloud

Abstract

We have carried out near-infrared spectroscopic observations of 23 very low-luminosity young stellar object (YSO) candidates and 5 their companions in Heiles Cloud 2, one of the densest parts of the Taurus molecular cloud. Twelve objects were confirmed as YSOs by Br gamma feature. The effective temperatures of the YSOs and of the companions are estimated from the 2.26 micron feature, the 2.21 micron feature, and the H2O band strengths. Detailed comparisons of our photometric and spectroscopic observations with evolutionary tracks on the HR diagram suggest some objects to be very low-mass YSOs.

Full text

Page 1

arXiv:astro-ph/0206047v1 4 Jun 2002
PASJ: Publ. Astron. Soc. Japan , 1–??,
c ? 2008. Astronomical Society of Japan.
Near-Infrared Spectroscopy of Very Low-Luminosity Young Stellar
Objects in the Taurus Molecular Cloud
Yoichi Itoh
Graduate School of Science and Technology, Kobe University, 1-1 Rokkodai, Nada, Kobe, Hyogo 657-8501
yitoh@kobe-u.ac.jp
Motohide Tamura
National Astronomical Observatory of Japan, 2-21-1 Osawa, Mitaka, Tokyo 181-8588
tamuramt@cc.nao.ac.jp
and
Alan T. Tokunaga
Institute for Astronomy, University of Hawaii 2680 Woodlawn Drive, Honolulu, Hawaii 96822, USA
tokunaga@ifa.hawaii.edu
(Received 2000 December 31; accepted 2001 January 1)
Abstract
We have carried out near-infrared spectroscopic observations of 23 very low-luminosity young stellar
object (YSO) candidates and 5 their companions in Heiles Cloud 2, one of the densest parts of the Taurus
molecular cloud. Twelve objects were confirmed as YSOs by Brγ feature. The effective temperatures of the
YSOs and of the companions are estimated from the 2.26 µm feature, the 2.21 µm feature, and the H2O
band strengths. Detailed comparisons of our photometric and spectroscopic observations with evolutionary
tracks on the HR diagram suggest some objects to be very low-mass YSOs.
Key words: stars: formation — stars: low-mass stars, brown dwarf — infrared: stars
1. Introduction
Recent
studies
stellar
forming
al.(1993);
000
000
[cite]cite.Oasa99Oasa et
mass star forming regions (000 [cite]cite.AspinAspin
etal.(1994)),and even
ing regions (000 [cite]cite.KaifuKaifu
000[cite]cite.Lucas00Lucas
[cite]cite.Oasa02Oasa et al(2002)).
is suggestive of low-mass; they may be very low-mass
young stars near the stellar/substellar boundary, young
brown dwarfs, or even free floating planets.
Fromthephotometric
ever, it is impossible to simultaneously determine
the mass and age of a YSO. Near-infrared spec-
troscopy of YSO candidates is necessary to over-
comethis difficulty (000
& Meyer(1995);000
etal.(1997); 000 [cite]cite.LuhmanRiekeLuhman
& Rieke(1998); 000[cite]cite.Luhman98Luhman
al.(1998); 000 [cite]cite.WilkingWilking et al.(1999);
000 [cite]cite.CushingCushing
[cite]cite.Lucas01Lucas et al.(2001)).
faint YSO candidates exhibit the absorption features
distinctive to late spectral type, implying young brown
optical
have
object
regions
000
[cite]cite.ITGItoh
[cite]cite.BarsonyBarsony
andnear-infrared
faintpopulations
candidates
[cite]cite.ComeronComeron
[cite]cite.Strom95Strom
et al.(1996),
et
al.(1999)),
photometric
of
low-mass
revealed
(YSO)
(000
young
starin
et
etal.(1995);
ITG;hereafter
al.(1997);
inintermediate-
000
in high-massstarform-
et al.(2000);
& Roche(2000);
Their faintness
000
observations alone, how-
[cite]cite.Greene95Greene
[cite]cite.Luhman97Luhman
et
et al.(2000);
Spectra of some
000
dwarfs.
ubiquitous in star-forming regions, detail of the formation
process of such objects, for example mass function, object
density, and disk property, is still unknown.
te]cite.Briceno98Brice˜ no(1998) ([cite]cite.Briceno98Brice˜ no
(1998)) have carried out an optical search for very low-
mass YSOs in the L1495, L1529, L1551, and B209
regions in the Taurus molecular cloud.
etry with spectroscopy, they found 9 new YSOs in
the clouds, half of them have very late spectral types,
implying very low-mass objects (0.05 M⊙– 0.25 M⊙).
te]cite.Luhman00Luhman(2000) ([cite]cite.Luhman00Luhman(2000))
further investigated very low-mass YSOs in the same
clouds by combining optical imaging with near-infrared
spectroscopy. They found that the mass function of
the YSOs in these regions has a peak at 0.8 M⊙and
is relatively flat between 0.1 M⊙and 0.8 M⊙range.
In these papers, however, because the targets for the
spectroscopy were fully or partially selected based on the
optical color-magnitude diagram, the sample may not be
complete especially for embedded very low-mass objects.
ITG conducted a near-infrared survey of the central 1◦×
1◦region of Heiles Cloud 2 in the Taurus molecular cloud,
one of the best-studied low-mass star forming regions,
with a limiting magnitude of 13.4 mag in the K-band.
Fifty YSO candidates were identified by their intrinsic
red color on the (J − H, H − K) color-color diagram,
following the scheme discussed by te]cite.Strom93Strom
etal.(1993)([cite]cite.Strom93Strom
Successive high-resolution imaging survey discovered
5 companion candidates around the YSO candidates
While such very low-mass objects may be
From photom-
etal.(1993)).

Page 2

2Itoh et al. [Vol. ,
(000 [cite]cite.ITNItoh et al.(1999)). Faintness of some
YSOs and their companion may be an indicative of the
low-mass of the objects.
We describe here near-infrared spectroscopic follow-up
of these faint YSO candidates in the Heiles Cloud 2. The
observations are described in §2, and data reduction pro-
cedures in §3. In §4, we derive effective temperatures of
the YSOs mainly from the 2.21 µm feature, the 2.26 µm
feature, and the H2O absorption band. We also calculate
photospheric luminosity of the YSOs from the previous
photometry, then plot the YSOs on the HR diagram.
2. Observations
2.1. UKIRT observations
Twenty-one YSO candidates were selected for the spec-
troscopic observations (table 1).
flux-limited; The K-band magnitudes range from 8 to 12,
up to 4 mag fainter than that of classical T Tauri Stars
(CTTSs) in the Taurus molecular cloud. We missed ITG
8 and ITG 34 probably due to errors in the coordinates.
As template stars, 3 late-type CTTSs, 1 class I source,
and 7 late-type dwarfs were also observed.
Spectroscopic observations have been carried out dur-
ing 1996 November 24 – 26 using the UK Infrared
Telescope (UKIRT) at the summit of Mauna Kea,
with the Cooled Grating Spectrometer 4 (CGS4, 000
[cite]cite.MountainMountain et al.(1991)). CGS4 has a
256×256 InSb array with a spatial scale of 1.′′22 pixel−1.
The 75 line mm−1grating was used with the 1.′′2 slit,
providing wavelength coverage of 1.8 µm – 2.5 µm with
resolution R(=λ/∆λ) of 300 at 2.21 µm. The integration
time used for each source was typically 1 – 20 seconds de-
pending on source brightness. In most cases, 96 exposures
were taken for each object. Nodding of the telescope was
carried out approximately 16 arcsec along the slit for sky
subtraction. For the YSO candidates and CTTSs, SAO
76542 (A2V) was observed for correction of the effects
of telluric absorption. For late-type dwarfs, we observed
stars with spectral types of A0V – A3V at similar air-
mass. Exposures of an incandescent lamp on and off were
taken at the start of each night as dome flats. Exposures
of a krypton lamp were taken every three or four hours
for wavelength calibrations.
The sample is nearly
2.2. Subaru observations
The observed objects are 5 binary candidates listed
in te]cite.ITNItoh et al.(1999) ([cite]cite.ITNItoh et
al.(1999)) (table 1). The K-band magnitudes of the com-
panions range from 13 to 16. Six late-type dwarfs were
also observed as templates, among which 3 latest dwarfs
were observed both with UKIRT and Subaru.
Spectroscopic observations have been carried out on
2000 December 4 – 5 with the Infrared Camera and
Spectrograph (IRCS, 000 [cite]cite.Tokunaga98Tokunaga
etal.(1998);000 [cite]cite.KobayashiKobayashi
al.(2000)) on the Subaru Telescope at the summit of
Mauna Kea, Hawaii. IRCS has a 1024×1024 InSb array
with spatial scale of 0.′′058 pixel−1. Typical seeing size
et
was 0.′′4 with a stable condition for both observing dates,
so that all binaries were well separated.
provided a wavelength coverage of 2.0 µm – 2.5 µm. The
resolving power R was around 350 at 2.2 µm. The slit
width we used is 0.′′6, slightly wider than the seeing size,
so that the effective spectral resolution might change
by seeing size. We measured FWHMs of the absorption
features in each spectrum for each object, and confirmed
that the FWHMs do not change in 1 pixel resolution
(∼ 6˚ A). Therefore, the effective spectral resolution was
stable during observations of each target. The integration
time used for each source was typically 60 – 300 seconds
depending on source brightness.
were taken for each object with the telescope dithered
approximately 7.′′5 along the slit for sky subtraction.
For the binary candidates, SAO 76542 was observed
for correction of the effects of telluric absorption. For
late-type dwarfs, we observed stars with spectral types
of A0V – A3V at similar airmass.
incandescent lamp on and off were taken at the end of
each night as dome flats. Exposures of an argon lamp
were taken for wavelength calibrations at the end of each
night.
The grating
8 or 12 exposures
Exposures of an
3. Data Reduction and Results
The Image Reduction and Analysis Facility (IRAF)
software was used for all data reduction. Reduction pro-
cedure is similar for both UKIRT and Subaru data. First,
a dithered pair of object frames were subtracted by each
other, then divided by flat fields. Next, the data frames
were geometrically transformed to correct the curvature
of the slit image caused by the grating. The solutions of
the geometric transformation and the wavelength calibra-
tion were derived from the spectrum of a krypton lamp or
argon lamp taken closest in time to the object. Then, in-
dividual spectra were extracted from the transformed im-
ages using the APALL task. The region where the inten-
sity of the object was more than 20 % of the peak intensity
at each wavelength was summed into a one-dimensional
spectrum. Extracted spectra were then normalized and
combined to produce the final spectra.
noise spectra due to the tracking error of the telescope
were rejected. The object spectrum was divided by the
standard star spectrum, in which Brackettγ absorption
line at 2.166 µm was removed by interpolating across the
adjacent continuum with the SPLOT task. Finally the
spectrum was multiplied by a blackbody spectrum of the
temperature appropriate to the spectral type of the stan-
dard star (000 [cite]cite.TokunagaTokunaga(2000)).
The K-band spectra of the 23 YSO candidates, 5 com-
panion candidates, 3 CTTSs, a protostar, and 10 late-type
dwarfs are shown in figures 1 and 2. Prominent features
in the spectra are the HI Brγ (2.17 µm), 2.21 µm feature,
2.26 µm feature, CO band (2.29 µm and longer), and H2O
absorption (<2.15 µm and >2.3 µm).
Equivalent widths of the Brγ line, the 2.21 µm feature,
and the 2.26 µm feature were measured with SPLOT task
by Gaussian fitting. On the other hand, equivalent widths
Low signal-to-

Page 3

No. ]Taurus YSOs3
Table 1. The YSO sample in Heiles Cloud 2
ITG No.
2
4
5
6
9A
9B
13
15A
15B
17
18
21
24
25A
25B
27
28
29
33A
33B
36
39
40
41
43
45A
45B
46
α(1950)
4h34m57s0
4 35 15.0
4 35 17.1
4 35 17.5
4 35 56.6
4 35 56.6
4 36 15.2
4 36 40.5
4 36 40.4
4 36 43.1
4 36 46.3
4 36 57.5
4 37 03.6
4 37 03.8
4 37 03.6
4 37 32.7
4 37 46.2
4 37 52.5
4 38 04.2
4 38 04.1
4 38 10.0
4 38 16.4
4 38 20.8
4 38 34.6
4 38 39.8
4 38 45.3
4 38 45.2
4 38 52.3
δ(1950)
25◦53′01′′
26 01 28
25 46 06
26 03 19
25 27 17
25 27 21
25 31 44
25 56 02
25 56 06
25 55 49
25 41 02
25 50 38
26 01 26
25 59 38
25 59 36
25 52 34
25 13 16
25 16 19
25 50 23
25 50 28
25 12 27
25 28 24
25 38 10
25 50 46
25 27 30
25 42 37
25 42 37
25 47 16
K J-H H-KAv Obs.∗
U
U
U
U
S
S
U
U,S
S
U
U
U
U
U,S
S
U
U
U
U,S
S
U
U
U
U
U
S
S
U
Identification†
10.05±0.01
11.05±0.02
8.31±0.00
10.37±0.01
12.53±0.03
14.45±0.03
11.69±0.02
9.02±0.00
14.50±0.01
10.37±0.01
10.27±0.01
10.83±0.01
13.05±0.05
8.97±0.00
13.38±0.01
8.46±0.00
11.87±0.02
10.74±0.01
11.50±0.02
16.05±0.02
11.01±0.02
11.80±0.02
11.47±0.02
9.72±0.01
11.24±0.01
12.77±0.05
15.76±0.05
11.01±0.01
0.84±0.01
2.07±0.05
1.17±0.01
0.96±0.02
0.88±0.04
0.70±0.04
0.84±0.03
0.86±0.01
1.01±0.02
0.96±0.02
1.96±0.03
1.49±0.03
1.01±0.09
1.97±0.01
2.14±0.01
1.04±0.00
0.81±0.04
0.86±0.01
1.41±0.04
0.79±0.04
0.86±0.02
1.01±0.03
2.80±0.18
1.66±0.01
0.84±0.01
0.90±0.06
1.33±0.08
0.83±0.01
0.60±0.01
1.29±0.03
0.72±0.01
0.87±0.02
0.66±0.04
0.70±0.04
0.55±0.03
0.62±0.01
0.73±0.02
0.82±0.02
1.18±0.02
0.94±0.02
0.70±0.09
1.37±0.01
1.28±0.01
0.64±0.00
0.53±0.04
0.54±0.01
1.19±0.03
0.54±0.03
0.54±0.02
0.66±0.03
1.99±0.05
1.11±0.01
0.53±0.01
0.73±0.07
0.91±0.09
0.52±0.01
0.14±0.49
11.42±0.55
3.38±0.47
0.14±0.49
0.27±0.42
0.00±0.39
0.36±0.51
0.19±0.47
1.48±0.39
0.45±0.49
10.6±0.50
6.21±0.51
1.65±0.82
9.82±0.49
12.29±0.38
2.28±0.47
0.09±0.54
0.63±0.49
3.94±0.55
0.00±0.41
0.63±0.49
1.88±0.51
16.68±0.83
7.38±0.49
2.97±0.55
0.21±0.52
4.42±0.62
0.41±0.49
Kim 1-19
GM Tau
separation = 4.′′31
separation = 2.′′99
GKH 5
IRAS 04370+2559
separation = 4.′′29
separation = 5.′′17
GKH 32
IRAS 04385+2550
separation = 2.′′29
∗U: UKIRT, S: Subaru
†GKH: te]cite.GomezGomez et al.(1994) ([cite]cite.GomezGomez et al.(1994)), Kim: te]cite.KimKim(1990) ([cite]cite.KimKim(1990))
of the CO(2-0) band and the CO(4-2) band were calcu-
lated by simple integration of the absorption intensity.
The uncertainties were estimated from the continuum fit,
in which locating the continuum level was the main factor
contributing to the equivalent width uncertainty.
also calculated reddening-independent indices of the H2O
bandstrengthQ, following
etal.(1999) ([cite]cite.WilkingWilking
al.(1999)).ForKoornneef’s
[cite]cite.KoornneefKoornneef(1983)),
written as
?1.24
We
te]cite.WilkingWilking
et
extinction
the Q index is
law(000
Q =
?F1
F2
??F3
F2
(1)
where F1, F2, and F3are flux densities between 2.07 µm
– 2.13 µm, 2.267 µm – 2.285 µm, and 2.40 µm – 2.45 µm,
respectively.
The measured equivalent widths of the features and
the strengths of the bands are tabulated in table 2. The
signal-to-noise ratio derived from deviations between each
exposure is typically 60 (table 2). For the latest dwarfs
taken both with UKIRT and Subaru, shapes of the spectra
are similar for both observations and most of the measured
equivalent widths and band strengths are within observa-
tional uncertainties.
4. Discussion
4.1.Identification as YSOs using the Brγ feature
We identified 12 sources as YSOs, whose Brγ feature is
in emission or in flat (featureless). We regard a feature-
less spectrum around the Brγ line with uncertainty less
than 2˚ A in equivalent width as flat. The line emission is
probably due to accretion of matter from the circumstel-
lar disk onto the star (e.g. 000 [cite]cite.NajitaNajita et
al.(1996)).
The other 11 objects with Brγ absorption are, on
the other hand, likely early-type field-stars. Equivalent
widths of the Brγ absorption feature are 4˚ A– 15˚ A,
consistent with the spectral types of B, A, and early F
(000 [cite]cite.AliAli et al.(1995)).
the color-color diagram are close to that of early-type
field stars. Figure 3 shows the color-color diagram of
the YSO candidates and field stars identified by ITG
toward Heiles Cloud 2.The objects with the Brγ ab-
sorption tend to be located near the boundary between
Their locations on

Page 4

4 Itoh et al. [Vol. ,
Table 2. Equivalent widths of the features and strengths of the band of the YSO candidates, the CTTSs, and the M dwarfs.
Object Brγ
[˚ A]
2.21 µm 2.26 µm CO(2-0)CO(4-2)Q Obs.∗
S/N
[˚ A][˚ A][˚ A]
6.20
···
2.71
2.30
3.45
2.71
···
6.73
5.88
25.90
6.79
[˚ A]
7.80
···
1.97
3.79
4.72
2.35
···
10.39
9.79
26.40
10.02
ITG 2
ITG 4
ITG 5
ITG 6
ITG 9A
ITG 9B
ITG 13
ITG 15A&B
ITG 15A
ITG 15B
ITG 17
ITG 18
ITG 21
ITG 24
ITG 25A&B
ITG 25A
ITG 25B
ITG 27
ITG 28
ITG 29
ITG 33A&B
ITG 33A
ITG 33B
ITG 36
ITG 39
ITG 40
ITG 41
ITG 43
ITG 45A
ITG 45B
ITG 46
DD Tau (M1)
GH Tau (M2)
FP Tau (M5.5)
IRAS 04365+2535
GJ 380 (K5)
GJ 96 (M1)
GJ 205 (M1.5)
GJ 85.1 (M3)
GJ 806 (M3)
GJ 251 (M3)
GJ 12 (M5)
GJ 406 (M6)
GJ 406 (M6)
LHS 2397a (M8)
LHS 2397a (M8)
BRI 0021-214 (M9.5)
BRI 0021-214 (M9.5)
0.0±0.8
6.80±0.25
4.24±1.50
-2.25±0.62
0.0±0.5
2.18±0.39
0.0±0.3
0.80±0.26
0.98±0.23
2.42±0.04
1.00±0.16
0.0±0.1
1.44±0.28
0.53±0.22
2.38±0.08
0.0±0.2
0.0±0.4
4.12±1.24
3.19±0.30
14.19±12.37
2.48±0.17
0.0±0.3
2.31±0.44
0.58±0.12
2.54±0.50
2.26±0.08
2.53±1.13
1.33±0.70
0.0±0.4
0.0±0.1
3.76±0.37
3.92±0.47
0.51±0.00
0.98±0.01
0.83±0.01
0.81±0.01
0.89±0.02
0.88±0.09
0.94±0.01
0.64±0.01
0.61±0.00
1.65+1.41
−0.82
0.57±0.01
0.98±0.01
0.54±0.01
1.03±0.01
0.87±0.03
0.89±0.02
0.73±0.09
0.86±0.01
1.00±0.01
0.96±0.01
0.74±0.01
0.73±0.01
0.90±0.04
0.91±0.01
1.00±0.02
0.67±0.02
0.93±0.01
0.97±0.01
0.88±0.07
1.55+1.77
−0.83
0.80±0.01
0.82±0.01
0.85±0.01
0.75±0.00
1.06±0.02
0.93±0.01
0.84±0.01
0.87±0.01
0.66±0.01
0.81±0.01
0.77±0.01
0.71±0.01
0.55±0.01
0.60±0.00
0.51±0.01
0.51±0.01
0.52±0.01
0.49±0.01
U
U
U
U
S
S
U
U
S
S
U
U
U
U
U
S
S
U
U
U
U
S
S
U
U
U
U
U
S
S
U
U
U
U
U
S
U
S
U
U
S
U
U
S
U
S
U
S
70
90
90
90
300
60
60
80
460
0±30±3
14.08±0.70
-0.79±0.79
-3.93±1.08
0.0±2.5
0.0±0.5
8.25±0.62
-3.8±3.8
6.60±1.69
-5.34±0.91
-12.63±0.22
0.0±3.4
0.0±1.2
5.52±0.72
11.54±0.80
-2.83±1.05
-5.85±0.85
0.0±2.9
5.14±1.63
7.40±0.70
0.0±0.9
0.0±1.0
7.68±0.98
8.26±0.96
0.0±0.3
3.29±1.00
3.29±0.24
3.53±2.92
1.96±0.37
0.0±0.3
2.22±0.78
0.0±0.5
1.94±0.51
3.38±0.16
1.41±0.60
1.36±0.64
0.75±0.15
0.47±0.12
3.68±0.43
3.78±0.54
7.54±4.90
1.20±0.43
0.0±0.2
4.15±1.00
2.31±0.50
0.0±0.4
0.0±0.88
9
30
80
70
40
··· ···
5.11
···
3.72
3.52
9.61
7.87
···
···
6.16
6.02
0±9
···
···
4.95
3.43
···
···
···
11.13
3.21
5.01
6.91
8.50
···
5.43
8.26
21.31
8.50
100
290
60
100
50
65
55
660
20
50
60
30
100
60
80
···
···
8.56
7.35
···
···
···
10.76
6.53
···
1.26±0.33
0.0±0.2
1.78±0.53
2.08±0.56
0.0±0.2
0.0±1.8
···
···
···
0±25
0.0±1.0
-2.53±0.35
-0.58±0.57
-1.02±0.46
-1.68±0.94
-0.14±0.25
-1.80±1.80
0.00±0.47
-1.04±0.74
-0.40±0.40
0.0±0.1
0.0±0.7
-1.72±0.76
0.0±0.1
0.0±0.7
-2.10±1.19
0.0±0.3
0.0±0.6
0±40±9
6
1.46±0.50
1.99±0.42
2.91±0.46
1.34±0.33
1.84±0.25
2.81±0.48
3.47±0.38
0.0±0.2
4.14±0.25
4.24±0.24
5.80±0.30
1.97±0.39
3.97±0.36
4.01±0.20
3.37±0.27
1.65±0.15
2.10±0.95
0.39±0.07
0.24±0.24
1.75±0.82
0.05±0.02
13.69
4.93
6.92
8.27
70
100
130
150
160
630
110
780
310
100
620
60
60
510
90
420
70
200
±0.013.35±0.69
0.0±0.2
4.20±0.19
4.78±1.18
7.16±0.51
2.60±0.60
4.08±0.34
5.67±0.47
4.39±0.34
7.67±0.75
7.23±0.31
4.98±0.31
5.04±0.15
2.87±0.52
2.98±0.26
······
6.28
6.67
4.37
3.74
4.88
4.90
···
8.13
6.02
14.32
11.44
11.54
9.57
9.56
7.19
8.31
5.11
7.39
8.08
···
7.96
7.35
13.54
12.79
10.59
7.60
∗U: UKIRT, S: Subaru

Page 5

No. ]Taurus YSOs5
Fig. 1. (a) K-band spectra of the YSO candidates taken with
UKIRT. The spectra are normalized by the flux between 2.18
µm and 2.20 µm, then offset in steps of 0.5.
Fig. 1. (b) K-band spectra of the YSO candidates taken
with UKIRT. The spectra are offset in steps of 0.5.
Fig. 1. (c) K-band spectra of the YSO candidates taken with
UKIRT. The spectra are offset in steps of 0.5.
Fig. 1. (d) K-band spectra of the YSO binary candidates
taken with Subaru. Additive constants for the spectra are 0,
1, 1.5, and 2, respectively.