THE ASTROPHYSICAL JOURNAL, 486:L141L144, 1997 September 10
©1997. The American Astronomical Society. All rights reserved. Printed in U.S.A.

---

Up:

Interferometric ¹³CO Observations of the Cometary Bright-rimmed Cloud 37 in IC 1396: Evidence for Collapse Phase of Radiation-driven Implosion?

KOJI SUGITANI

Institute of Natural Sciences, Nagoya City University, Mizuho-ku, Nagoya 467, Japan

KOH-ICHIRO MORITA

Nobeyama Radio Observatory, Minamimaki, Minamisaku, Nagano 384-13, Japan

MAKOTO NAKANO

Faculty of Education, Oita University, Oita 870-11, Japan

MOTOHIDE TAMURA

National Astronomical Observatory of Japan, Osawa 2-21-1, Mitaka, Tokyo 181, Japan

AND

KATSUO OGURA

Kokugakuin University, Higashi, Shibuya-ku, Tokyo 150, Japan

Received 1997 March 10; accepted 1997 June 30

ABSTRACT

     Interferometric observations of the cometary bright-rimmed cloud BRC 37 (GN 21.38.9) located at the periphery of the H II region IC 1396 were made with the Nobeyama Millimeter Array in the ¹³CO (J = 10) transition. Although the global distribution of the molecular gas is consistent with the previous single-dish results, our high-resolution data clearly show small-scale structures with blueshifted velocity components corresponding to two clear tails stretching from the globule head in the opposite direction toward the exciting star of IC 1396. The elongation direction of these tails and other circumstances suggest an interaction with the UV radiation from the exciting star of IC 1396. The position-velocity diagram along the cloud axis shows a velocity pattern which may be explained by the collapsing gas motion. The morphology as well as the position-velocity diagram seems to be in agreement with the collapse phase of the recent two-dimensional simulations of the radiation-driven implosion of the neutral globule due to UV radiation from OB stars.

Subject headings: H II regionsISM: globulesISM: individual (IC 1396)radio lines: ISMstars: formationstars: premain-sequence

CONTENTS

• 1. INTRODUCTION
• 2. OBSERVATIONS AND DATA REDUCTION
• 3. RESULTS
• 4. DISCUSSION
• ACKNOWLEDGMENTS
• REFERENCES
• FIGURES
• REFERENCES TO THIS ARTICLE

§1. INTRODUCTION

     Since Reipurth (1983) reported the possibility of star formation in CG 1, small cometary globules with bright rims have been considered to be one of potential sites of star formation. In IC 1396, which is an extended old H II region at the distance of 750 pc (e.g., Becker & Fenkart 1971; Battinelli & Capuzzo-Dolcetta 1991), there are many globules surrounded by bright rims due to UV radiation from the exciting star. They have been the focus of many studies (e.g., Weikard et al. 1996 and references therein), and in particular the triggering of star formation by UV radiation has been discussed (e.g., Sugitani et al. 1989; Nakano et al. 1989; Duvert et al. 1990; Kun & Pasztor 1990; Seraby, Gusten, & Mundy 1993; Patel et al. 1995; Weikard et al. 1996).

     BRC 37 is a small cometary globule with a bright rim located in the southern periphery of IC 1396, which is also identified as GRS 12 in Gyulbudaghian (1985), GN 21.38.9 in Duvert et al. (1990), or rim J in Weikard et al. (1996). The size of the globule is about 1 wide and 5 long in the optical. IRAS 21388+5622 is located at the head part of the globule and has an IRAS co-added luminosity of 155 L, which corresponds to that of an A star with far-IR colors of class II sources (Schwartz, Gyulbudaghian, & Wilking 1993). Preliminary results of our recent 2 mm continuum observations of BRC 37 with a spatial resolution of 12 (Sugitani et al. 1997a) suggest that a 2 mm source corresponds to IRAS 21388+5622 with a position of (R.A.[1950] = 21^h38^m5290 ± 045, decl.[1950] = 56°22165 ± 39) and with a flux of 122 ± 44 mJy. Duvert et al. (1990) has extensively studied the globule with the IRAM telescope in the CO (J = 10), ¹³CO (J = 10), and CS (J = 21) transitions. By discovering a bipolar molecular outflow associated with the IRAS point source, they showed that star formation was taking place there. They also pointed out the possibility that this star formation was induced by the interaction with the strong UV field from the O6 star HD 206267, based on the fact that the dense head core of the cometary globule, covered by the ionization front, faced toward this O6 star.

     Sugitani et al. (1991) and Sugitani & Ogura (1994) have made systematic surveys of bright-rimmed clouds/globules associated with IRAS point sources in a search for candidates of recent star formation induced by UV radiation around H II regions by using the Palomar Observatory Sky Survey (POSS) prints in the northern sky and the ESO (R)/SERC (J) atlases in the southern sky. As a result, they found in total 89 such objects in the whole sky, including BRC 37, which is listed as cloud 37 in their catalog. For 44 objects of the northern sky, Sugitani, Tamura, & Ogura (1997b) have conducted JHK imaging and found many small clusters of near-infrared sources having young stellar object (YSO) colors. Very interestingly, in at least six bright-rimmed clouds the clusters are found to have elongations of 13 toward the exciting star(s) of the bright rims, with the IRAS sources situated near the other end. (The positional errors of the IRAS sources are 830.) In addition, there is a tendency for bluer (i.e., older) stars to be located closer to the exciting star(s) and redder (i.e., younger) stars closer to the IRAS sources. They concluded that this asymmetric distribution of the cluster members strongly suggested small-scale sequential star formation, i.e., a propagation of star formation from the side of the exciting star(s) to the IRAS position due to the advance of the ionization/shock front caused by the UV radiation from the exciting star(s) (Sugitani, Tamura, & Ogura 1995). BRC 37 is one of the examples of such sequential star formation.

     The interaction between cometary globules and UV radiation from OB star(s) has been studied as one possible mechanism for star formation, which is called radiation-driven implosion (RDI) (e.g., Sandford, Whitaker, & Klein 1982; Bertoldi 1989; Elmegreen 1993 and references therein). Recently, Lefloch & Lazareff (1994) presented their results for the two-dimensional hydrodynamical simulations concerning the dynamical evolution of a neutral globule illuminated by UV radiation of OB stars. They showed that RDI is composed of two stages: the first stage is a collapse phase of a short timescale of a few times 10⁵ yr, and the second is an equilibrium cometary phase of a few times 10⁶ yr. This two-staged process was also shown by using analytical treatments of RDI (Bertoldi 1989; Bertoldi & McKee 1990). Lefloch & Lazareff (1995) have made a comparison of their single-dish observations of a globule in IC 1848 with the numerical simulations and reported that the observational data were consistent with the interpretation that the globule was in the transient phase between the two stages.

     In this Letter we report the results of observations of the bright-rimmed cometary globule BRC 37 (GN 21.38.9) in IC 1396 in the ¹³CO (J = 10) transition obtained by the Nobeyama Millimeter Array (NMA), and try to compare them with the results of the simulations by Lefloch & Lazareff (1994).

§2. OBSERVATIONS AND DATA REDUCTION

     Observations of the ¹³CO (J = 10) line were made with NMA on 1995 November 25 and 26. We used the D-array configuration of six 10 m dishes. The minimum separation of the telescopes was 15 m, and the maximum was 70 m. The front ends were SIS receivers, and their system temperature was typically 300 K during the observations. The back end was the 1024 channel digital spectrometer with a bandwidth of 80 MHz, of which the velocity resolution was 0.21 km s^-1. The center of the field of view was the position of IRAS 21388+5622 (R.A. = 21^h38^m532, decl. = 56°2218 [1950]). The phase and flux calibration were made by observing BL Lac every 30 minutes, and the bandpass was calibrated by observing 3C 273. The absolute flux calibration was made by employing Uranus, assuming its surface temperature of 120 K, and the BL Lac flux of 4.8 Jy at 110 GHz was obtained. The uv data were processed by the standard procedure and CLEANed to obtain nonweighted channel maps and position-velocity diagrams by using AIPS. The synthesized beam was 76 × 47 with P.A. 1714. The primary-beam size of the 10 m telescopes was about 1 at 3 mm, which covers only the head part of the cometary globule. Line-free channels were used for mapping the continuum emission. The typical rms noise levels of a single channel and the continuum map were 0.24 and 0.0082 Jy beam^-1, respectively.

§3. RESULTS

     Figures 1a1n (Plate L19) show the channel maps of the LSR central velocities from 1.91 km s^-1 down to -0.85 km s^-1 with an interval of 0.21 km s^-1. The redshifted emission found in the maps between 1.91 and 1.27 km s^-1 is concentrated mostly in the vicinity of the IRAS source. As the velocity decreases, the emission becomes more extended and elongated in the direction of the cometary tail, although the intensity peak is still in the vicinity of the IRAS source. In the maps of 0.21 km s^-1, the emission splits into two elongated components. Similar structures of larger scales were reported in BRC 38 (rim E) of IC 1396 by Patel et al. (1995) and in the massive cometary globule BRC 7 of IC 1805 (Heyer et al. 1996). Here we call these elongated components wings. The wings stretch from the head of the globule in the direction opposite to the exciting star. Figure 2 (Plate L20) illustrates the three-channel maps of 1.49, 0.64, and -0.21 km s^-1 overlaid on the optical image from the Digitized Sky Survey, clearly showing the general trend of the velocity structure mentioned above. The (0, 0) position is that of IRAS 21388+5622. Only a near-infrared source (source A of Fig. 1o) having an elongated reflection nebula is located within the positional error ellipse of the IRAS point source (semimajor axis, 8, semiminor axis, 6, and P.A. 179°). Other sources, B and C, are just outside the error ellipse. Our 2 mm observations also suggest that the IRAS counterpart is most likely source A (Sugitani et al. 1997a). This source is classified as a class I source based on the J - H/H - K color-color diagram (see, e.g., Itoh, Tamura, & Gatley 1996).

Fig. 1

Fig. 2

     Our ¹³CO (J = 10) observation has a field of view (FOV) of about 1, covering only the head part of the cometary globule, and lacks the zero-spacing data. Therefore, the single-dish observations executed by Duvert et al. (1990) with the IRAM 30 m telescope are very useful to compensate for the lack of the zero-spacing data. The global distribution of the ¹³CO gas in the head part of the globule is mostly consistent between our interferometric observations and the single-dish data, except for the small-scale structures within our FOV. The difference found in the results from the interferometer and single-dish observations is the appearance of the blueshifted gas. At the blueshifted velocity of V_LSR = 0.0 km s^-1, our FOV covers 80% of the ¹³CO emission area of the single-dish observation, judged from Figure 4b of Duvert et al. (1990). Our data show two elongated lobes, i.e., the wings, while the single-dish ¹³CO data do not show such lobe structures. However, the ¹²CO (J = 10) map of V_LSR = -0.25 km s^-1 (Fig. 4a of Duvert et al. 1990) gives a hint of one elongated structure at the highest contour of the ¹²CO emission extended toward the tail side over our FOV; this might correspond to one of the wings. Taking into account the difference in the appearance of ¹³CO, and the ¹²CO extended emission, it is suggested that the wings are seen because of the rim-brightening effect at both edges of the thin layer which covers the cloud surface. This layer is probably extended toward the tail side over our FOV.

     In detail, the IRAS position corresponds to the intensity peaks in the channel maps of 1.06 and 1.27 km s^-1 only, while it is in the immediate vicinity of the intensity peaks in the maps of 0.43 km s^-1. In the maps of 1.49 km s^-1 the peaks are slightly shifted to the southeastern direction, and between 0.85 and 0.43 km s^-1 to the northwestern direction. This is also noticed by comparing the higher contours for the two velocities of 1.49 and 0.64 km s^-1 of Figure 2. These facts indicate that a dense core with a velocity gradient of the northwest to southeast direction is associated with the IRAS point source. The direction of the velocity gradient is nearly perpendicular to the elongation of the reflection nebula illuminated by the above-mentioned near-infrared source, which probably corresponds to the IRAS point source (see Fig. 1o). The axis of the bipolar molecular outflow associated with this IRAS source is also similar to the elongation direction of the reflection nebula (Duvert et al. 1990). These may indicate the existence of a rotating core or disk associated with the IRAS source, although it is not well resolved by the present observation.

     In the channel maps of 1.49, 1.27, and 1.06 km s^-1, a protrusion toward the northwest direction from the main cloud is clearly seen around the (-8, +4) position that is very close to a red near-infrared source showing a color index between class I and class II sources on the J - H/H - K color-color diagram (source B of Fig. 1o). In the optical this source is not visible while another source with a color index of class II source is visible in the immediate vicinity (source C). The protrusion can be also identified in the channel maps of between 0.85 and -0.21 km s^-1. These indicate the existence of another cloud condensation/core around the (-8, +4) position, closer to the exciting star and that the near-infrared source was born and is still embedded there.

§4. DISCUSSION

     The appearance of the wings and the elongation of the globule (Fig. 1) indicate that the symmetry axis of the head part of BRC 37 is P.A. = 150° or 330°. This direction is nearly the same as that toward the main exciting star of IC 1396 (P.A. 340°; cf. Fig. 2 of Weikard et al. 1996). This agreement and the fact that the densest part of the globule is facing toward the exciting star may be due to the interaction between the molecular cloud and the UV radiation or stellar wind from the exciting star(s) of IC 1396. Patel et al. (1995) suggested that the total momentum of stellar wind in IC 1396 is not enough to explain the velocity shifts of the molecular clouds associated with IC 1396 and that the UV radiation is dominant in affecting the velocity structures of these clouds. Therefore, in the following we will discuss the detailed structures and evolution of BRC 37 by assuming that the main interaction agent is the UV radiation from the exciting star of IC 1396.

     Lefloch & Lazareff (1994) showed that, for a wide range of physical parameters including those in actual H II regions, RDI is a two-staged process; the first stage is a collapse with a short timescale of a few times 10⁵ yr, and the second is an equilibrium cometary phase of a few times 10⁶ yr. In the first stage, an isothermal shock progresses into the hemisphere of the globule on the OB star side, leaving compressed neutral gas behind, and forms a slightly elongated core on the symmetric axis of the globule near its original center/focus, and the compressed layer looks like a V or wing shape in the cross section (e.g., see Fig. 4 of Lefloch & Lazareff 1994). The compressed layer continues to progress toward the hemisphere of the rear (tail) side and then passes beyond the globule rear end, becoming protruded ears. The ears continue to develop and converge on the axis on the tail side. In the second stage, after the maximum compression, the globule overshoots the equilibrium state and reexpands. Afterward, a recompression occurs and the radial oscillation decreases its amplitude. Then, the globule reaches nearly a quasi-equilibrium of a cometary shape.

     Here we try to compare the ¹³CO data with the results of the simulations of Lefloch & Lazareff (1994) by assuming that RDI is taking place in BRC 37. In Figure 1 we can clearly identify a slightly elongated core/condensation of ¹³CO gas around the IRAS source and two wings/lobes which stretch from the vicinity of the IRAS source toward the globule tail with bluer velocities. Taking into account the positions and blueshifted velocities of these wings on the tail side as well as the possible projection effect, these wings may correspond to the compressed layer and/or the ears of Lefloch & Lazareff (1994). Therefore, the structure of the ¹³CO gas with the core and the two wings are considered to be consistent with the morphology expected in the collapse phase. We suspect that BRC 37 corresponds to a stage between two figures, Figures 4b and 4c of Lefloch & Lazareff (1994), judging from the morphology of the wings.

     We also try to explain the position-velocity diagram along the axis of the globule based on the Lefloch & Lazareff simulation. Figure 3 is a position-velocity diagram along the line of P.A. = 330° which passes the IRAS point source. A blueward velocity gradient toward the tail side of +0.02 km arcsec^-1 or 5.5 km s^-1 pc^-1 is seen throughout the head part. On the other hand, a redward velocity gradient of -0.2 km arcsec^-1 is seen near the IRAS position. The position-velocity diagrams along the symmetry axis at different times of the evolution in the simulations are shown in Figure 14 of Lefloch & Lazareff (1994). They mentioned that the position-velocity diagrams are not sensitive to the chosen parameters and that in the collapse phase the blueward velocity gradient along the symmetry axis and the redward velocity shift behind the head of the globule are seen for a globule inclined by 30° above the plane of the sky. The former is due to the gas accelerated through the tail, and the latter due to the front-side compressed/collapsing gas and the reexpanding rear-side gas, respectively. BRC 37 probably has an inclination angle smaller than 30°, judged from its global view, and the velocity gradient of 5.5 km s^-1 pc^-1 may be reconciled with 10 km s^-1 pc^-1 of the model in Lefloch & Lazareff (1994).

Fig. 3

     As mentioned in § 1, stars may have been formed sequentially from the side of the exciting star toward that of the IRAS point source in bright-rimmed globules, a phenomenon which we call small-scale sequential star formation (Sugitani et al. 1995). In Figures 1o and 1p, such a sequence is seen in a small cluster of six stars with JHK colors of YSOs. The IRAS source probably corresponds to a class I source with an elongated reflection nebula (source A of Fig. 1o), and the J - H/H - K color-color diagram suggests that the source associated with the protruded condensation (source B) is in the transient stage between class I and class II. One class II source (source C) is located just next to source B. Two class II sources (sources E and F), and one class III source (source D) that lies in the very vicinity of the class II area on the J - H/H - K diagram, are located outside the globule on the exciting-star side. The direction of the alignment of these six sources is nearly the same as that of the globule axis and toward the exciting star of IC 1396. Thus, the agreement of the directions also strongly suggests that star formation has been taking place in connection with the interaction between the cloud and the UV from the exciting star. The IRAS point source has the high luminosity of an intermediate-mass star (Schwartz, Gyulbudaghian, & Wilking 1993), and the other near-infrared sources have luminosities that correspond to low-mass (12 M) stars (Sugitani et al. 1995). Other bright-rimmed clouds that suggest such sequential star formation also show the same tendency.

     We think that there are two possible mechanisms of such sequential star formation. One is that within a single globule, star formation progresses continuously according to the movement of the densest part of the compressed gas along the symmetry axis from the exciting-star side to the tail side. Another is that, since a globule can be composed of multiple smaller condensations, the compression begins to occur in the condensation closest to the exciting star and then progresses step by step toward the tail side. Recently it has been suggested that molecular cloud cores have substructures of less than 0.1 pc (e.g., Langer et al. 1995). BRC 37 has the protruded condensation associated with the near-infrared source (source B) on the exciting-star side and some subcondensations on the tail side in addition to the core where the IRAS point source is located. The main core with the IRAS source has the largest size of about 0.1 pc, and the others have smaller sizes of less than 0.1 pc. In cloud 12 in the catalog of Sugitani et al. (1991) a small bright-rimmed condensation associated with a YSO is also seen toward the exciting-star side (see Fig. 3 of Sugitani et al. 1995). Thus, this evidence seems to be favorable for the latter mechanism. We speculate that the main core of BRC 37 has been formed by a merger of smaller condensations due to the focusing flow caused by the round surface of the cloud and, consequently, has become larger and so has formed a star of higher mass.

     Most of the previous studies concerning triggered star formation were based only on circumstantial evidence, and more direct evidence has long been desired. We consider that this observation is the first step in a more direct approach to triggered star formation that can be made with high-resolution facilities. It is, therefore, very important to make high-resolution observations of more bright-rimmed clouds or other candidate clouds of triggered star formation in order to clearly reveal the mechanism of such star formation. Particularly, observations in molecular lines of dipole moments higher than ¹³CO should be helpful for understanding the details of the structures and dynamics in such clouds.

ACKNOWLEDGMENTS

     We would like to thank Munetake Momose of the Nobeyama Radio Observatory (NRO) for helping with the data reductions, and other staff members and students of NRO for support in the observations. We would also like to thank the anonymous referee for useful comments. This work was financially supported in part by a Grant-in-Aid for Scientific Research by the Ministry of Education, Science, and Culture (08640332).

REFERENCES

• Battinelli, P., & Capuzzo-Dolcetta, R. 1991, MNRAS, 249, 76 First citation in article | NASA ADS
• Becker, W., & Fenkart, R. 1971, A&AS, 4, 241 First citation in article
• Bertoldi, F. 1989, ApJ, 346, 735 First citation in article | NASA ADS
• Bertoldi, F., & McKee, C. F. 1990, ApJ, 354, 529 First citation in article | NASA ADS
• Duvert, G., Cernicharo, J., Bachiller, R., & Gomez-Gonzalez, J. 1990, A&A, 233, 190 First citation in article | NASA ADS
• Elmegreen, B. G. 1993, Star Formation in Stellar Systems, ed. G. Tenorio-Tagle, M. Prieto, & F. Sánchez (Cambridge: Cambridge Univ. Press), 395 First citation in article
• Gyulbudaghian, A. L. 1985, Astrofizica, 23, 295 First citation in article | NASA ADS
• Heyer, M. H., et al. 1996, ApJ, 464, L175 First citation in article | Full Text | NASA ADS
• Itoh, Y., Tamura, M., & Gatley, I. 1996, ApJ, 465, L129 First citation in article | Full Text | NASA ADS
• Kun, M., & Pasztor, L. 1990, Ap&SS, 174, 13 First citation in article | NASA ADS
• Langer, W. D., Velusamy, T., Kuiper, T. B. H., Levin, S., Olsen, E., & Migenes, V. 1995, ApJ, 453, 293 First citation in article | NASA ADS
• Lefloch, B., & Lazareff, B. 1994, A&A, 289, 559 First citation in article | NASA ADS
• . 1995, A&A, 301, 522 First citation in article | NASA ADS
• Nakano, M., Tomita, Y., Ohtani, K., Ogura, K., & Sofue, Y. 1989, PASJ, 41, 1073 First citation in article | NASA ADS
• Patel, N. A., Goldsmith, P. F., Snell, R., Hezel, T., & Xie, T. 1995, ApJ, 447, 721 First citation in article | NASA ADS
• Reipurth, B. 1983, A&A, 117, 183 First citation in article | NASA ADS
• Sandford, M. T., Whitaker, R. W., & Klein, R. I. 1982, ApJ, 260, 183 First citation in article | NASA ADS
• Schwartz, R. D., Gyulbudaghian, A. L., & Wilking, B. A. 1993, 370, 263 First citation in article
• Seraby, E., Gusten, R., & Mundy, L. 1993, ApJ, 404, 247 First citation in article | NASA ADS
• Sugitani, K., Fukui, Y., Mizuno, A., & Ohashi, N. 1989, ApJ, 342, L87 First citation in article | NASA ADS
• Sugitani, K., Fukui, Y., & Ogura, K. 1991, ApJS, 77, 59 First citation in article | NASA ADS
• Sugitani, K., Matsuo, H., Nakano, M., Tamura, M., & Ogura, K. 1997a, in preparation First citation in article
• Sugitani, K., & Ogura, K. 1994, ApJS, 92, 163 First citation in article | NASA ADS
• Sugitani, K., Tamura, M., & Ogura, K. 1995, ApJ, 455, L39 First citation in article | Full Text | NASA ADS
• . 1997b, in preparation First citation in article
• Weikard, H., Wouterloot, J. G. A., Gastets, A., Winnewisser, G., & Sugitani, K. 1996, A&A, 309, 581 First citation in article | NASA ADS

FIGURES

Full image (156kb)

an

ao

Full image (88kb)

Full image (33kb)