1: % iaus2esa.tex -- sample pages for Proceedings IAU Symposium document class
2: % (based on v1.0 cca2esam.tex)
3: % v1.04 released 17 May 2004 by TechBooks
4: %% small changes and additions made by KAvdH/IAU 4 June 2004
5: % Copyright (2004) International Astronomical Union
6:
7: \NeedsTeXFormat{LaTeX2e}
8:
9: \documentclass{iaus}
10: \usepackage{graphicx}
11:
12: \title[Massive Star and Star Cluster Formation] %% give here short title %%
13: {Massive Star and Star Cluster Formation}
14:
15: \author[Tan] %% give here short author list %%
16: {Jonathan C. Tan$^1$}
17:
18: \affiliation{$^1$Department of Astronomy, University of Florida, Gainesville, FL 32611, USA
19: \break email: jt @ astro.ufl.edu}
20:
21: \pubyear{2006}
22: \volume{237} %% insert here IAU Symposium No.
23: \pagerange{???--???}
24: \date{?? and in revised form ??}
25: \setcounter{page}{1}
26: \jname{Triggered Star Formation in a Turbulent ISM}
27: \editors{B. G. Elmegreen \& J. Palous, eds.}
28: \begin{document}
29:
30: \maketitle
31:
32: \begin{abstract}
33: I review the status of massive star formation theories: accretion from
34: collapsing, massive, turbulent cores; competitive accretion; and
35: stellar collisions. I conclude the observational and theoretical
36: evidence favors the first of these models. I then discuss: the initial
37: conditions of star cluster formation as traced by infrared dark
38: clouds; the cluster formation timescale; and comparison of the initial
39: cluster mass function in different galactic environments.
40:
41: %\keywords{Keyword1, keyword2, keyword3, etc.}
42: %% add here a maximum of 10 keywords, to be taken form the file <Keywords.txt>
43: \end{abstract}
44:
45: \firstsection % if your document starts with a section,
46: % remove some space above using this command.
47: \section{Introduction}
48:
49: Massive stars and star clusters form together as part of a single
50: unified process. All locally-observed massive stars appear to form in
51: star clusters (de Wit et al. 2005), particularly in rich star clusters
52: (Massi, Testi \& Vanzi 2006). Star clusters make a significant,
53: perhaps dominant, contribution to the total star formation rate of
54: galaxies (Lada \& Lada 2003; Fall, Chandar, \& Whitmore 2005), so to
55: understand global star formation properties of galaxies
56: (e.g. Kennicutt 1998), one must understand star cluster formation.
57:
58: \section{Massive Star Formation}
59:
60: There is still some debate about how massive stars form. Do they form
61: from the global collapse of a massive, initially starless gas core, in
62: which a central protostar or binary grows from low to high mass by
63: accretion from a disk (e.g. McKee \& Tan 2003)? This is a
64: scaled-up version of the standard model of low-mass star formation
65: (Shu, Adams, \& Lizano 1987). Or do they form from favored low-mass protostellar
66: seeds that accrete gas competitively, with the gas being bound to the
67: protocluster potential but not at any stage in a spatially coherent
68: bound core with a mass similar to that of the final massive star
69: (e.g. Bonnell, Vine, \& Bate 2004). These latter models typically involve the
70: global collapse of the protocluster gas over a timescale
71: approximately equal to its free-fall time, so the growth of the
72: massive star takes place on the same timescale as the formation of the
73: entire cluster. It has been suggested that protostellar collisions may
74: also be involved in the growth of massive stars (Bonnell, Bate, \&
75: Zinnecker 1998; Bally \& Zinnecker 2005).
76:
77: Evidence in support of the core model of massive star formation
78: includes the fact that massive starless cores are observed and the
79: mass function of these cores is similar to the stellar initial mass
80: function (IMF) (Beuther \& Schilke 2004; Reid \& Wilson 2006).
81: Massive cores tend to have line widths that are much broader than
82: thermal (Caselli \& Myers 1995), indicating that other forms of
83: pressure support such as turbulent motions and magnetic fields are
84: important. Indeed observed magnetic field strengths are close to the
85: values needed to support the gas (Crutcher 2005). Known massive
86: protostars tend to be embedded in dense gas cores with masses
87: comparable to the stellar masses (e.g. Source I in the Orion Hot Core;
88: W3($\rm H_2O$)). Low-mass protostars, i.e. actively accreting stars,
89: always have relatively massive accretion disks and outflows. A number
90: of claims have been made for disks around massive protostars, although
91: it is usually difficult to determine if these are rotationally
92: supported structures (see Cesaroni et al. 2006 for a review). Powerful
93: outflows from massive protostars with similar degrees of collimation
94: to those from low-mass protostars have been seen (Beuther et
95: al. 2002). The expected evolutionary scheme for high-mass star
96: formation from cores has been reviewed in more detail by Beuther et
97: al. (2006). Doty, van Dishoeck, \& Tan (2006) considered the chemical
98: evolution of this model with particular application to observations of
99: water abundance in hot cores. Kratter \& Matzner (2006) investigated
100: the gravitational stability of massive protostellar accretion
101: disks. Krumholz, McKee, \& Klein (2007) presented
102: radiation-hydrodynamic simulations of massive star formation from a
103: massive turbulent core.
104:
105:
106:
107:
108: %One may ask why this mode of massive star formation is not evident in
109: %many of the numerical models of star formation, especially those based
110: %on the SPH technique. The answer is likely to be due to the lack of
111: %magnetic fields in these numerical models, which in real systems are
112: %observed to be dynamically important on a wide range of scales in
113: %star-forming regions (Crutcher 2007, these proceedings). Magnetic
114: %pressure dominates over thermal pressure on scales larger and more
115: %massive than the thermal Jeans mass, which is typically
116: %$<1\:M_\odot$. Thus the correct initial conditions for massive star
117: %formation can only arise when magnetic pressure is included. The mass
118: %function of starless cores is likely to be set by the fragmentation of
119: %a turbulent magnetized medium. Not all cores that form will be
120: %gravitationally unstable. The mass function of of cores is also likely
121: %to be affected by the merger and disruption of cores as they interact
122: %with one another. The cross-section for core-core interactions is
123: %relatively large: in the Turbulent Core Model (mcKee \& Tan 2003) the
124: %core is expected to interact with approximately its own mass of gas
125: %while it is undergoing collapse to a massive star. The fiducial core
126: %in this model is shown in Fig.~1.
127:
128:
129:
130:
131: Massive star formation models involving competitive accretion and
132: stellar collisions face several observational and theoretical hurdles.
133: Edgar \& Clarke (2004) showed that Bondi-Hoyle accretion becomes very
134: inefficient for protostellar masses $\gtrsim 10\:M_\odot$ because of
135: radiation pressure on dust in the gas. This feedback has not been
136: included in any of the simulations in which massive stars form by
137: competitive accretion.
138:
139: To overcome radiation pressure the accretion flow to a massive star
140: must become optically thick, either in a dense core or disk, or in
141: collisions of protostars. The collisional timescale is $t_{\rm
142: coll}=1.44\times 10^{10} (n_*/10^4{\rm pc^{-3}})^{-1} (\sigma/2{\rm
143: km\:s^{-1}}) (r_*/10R_\odot)^{-1} (m_*/M_\odot)^{-1}\:{\rm yr}$ in the
144: limit of strong gravitational focusing, where $\sigma$ is the 1D
145: velocity dispersion and $r_*$ is the radius of the stellar collisional
146: cross-section. For collisions to occur frequently enough to grow a
147: massive protostar within $10^6\:{\rm yr}$ (massive zero age main sequence
148: stars are observed) requires protostellar densities of at least
149: $10^6\:{\rm pc^{-3}}$ and probably closer to $10^8\:{\rm pc^{-3}}$,
150: whereas typical observed stellar densities around massive protostars
151: are much smaller. For example from the Orion Nebula Cluster (ONC) x-ray
152: observations of Garmire et al. (2000), Tan (2004) estimates a stellar
153: density of about $10^5\:{\rm pc^{-3}}$ in the KL region. This result
154: is not significantly changed by the deeper x-ray observations of
155: Grosso et al. (2005). Hunter et al. (2006) find a density of sub-mm
156: cores in the center of protoclusters in NGC~6334 of about $10^4\:{\rm
157: pc^{-3}}$. From Fig.~1 and the data of Mueller et al. (2002) we see
158: that typical mean densities of the central regions of Galactic
159: protoclusters are $n_{\rm H}\simeq 2 \times 10^5\:{\rm
160: cm^{-3}}$, i.e. 7000~$M_\odot\:{\rm pc^{-3}}$. If all this
161: gas mass formed stars, stellar
162: densities would be about $10^4\:{\rm pc^{-3}}$, given a typical
163: IMF. The fiducial core that forms a massive star in the model of McKee
164: \& Tan (2003) is also shown in Fig.~1, and has a mean density about
165: one ten times greater than this. Even if the core fragmented
166: with 100\% efficiency into low-mass stars the stellar density would be
167: too low for efficient growth via stellar collisions. In fact
168: numerical simulations show that fragmentation of the core into many
169: stars is impeded by heat input from the forming central massive
170: star (Krumholz 2006). The numerical simulations in which greater
171: degrees of fragmentation are seen (e.g. Dobbs, Bonnell, \& Clark 2005)
172: do not include this feedback. Magnetic pressure is also likely to be
173: important for the support of cores more massive than the thermal Jeans
174: mass, but this is also usually not included in simulations of massive
175: star formation.
176:
177: If collisions are relevant for massive star formation, but not
178: low-mass star formation, then one might expect a
179: change in the slope of the stellar IMF at the
180: mass scale at which the collisional process becomes important. In fact
181: the stellar IMF is reasonably well-fit by a power law from $\sim 1\:M_\odot$ out to the
182: highest observed masses (Massey 1998).
183:
184:
185:
186: \begin{figure}
187: \includegraphics[height=5.5in,width=5.5in,angle=0]{f1_tan.eps}
188: % \includegraphics{fig5prague.eps}
189: \caption{Surface density, $\Sigma\equiv M/(\pi R^2)$, versus mass,
190: $M$, for star clusters and interstellar clouds. Contours of constant
191: radius, $R$, and hydrogen number density, $n_{\rm H}$, or free-fall
192: timescale, $t_{\rm ff}$, are shown with dotted lines. The minimum
193: $\Sigma$ for CO clouds in the local Galactic FUV radiation field is
194: shown, as are typical GMC parameters and the distributions
195: $M(>\Sigma)$ of several IRDCs derived from extinction mapping
196: (Butler et al., in prep.). Open squares are star-forming clumps
197: (Mueller et al. 2002): a triangle indicates presence of an HII
198: region. The solid straight line traces conditions from the inner to
199: outer parts of the ONC, assuming equal mass in gas
200: and stars. Several more massive clusters are also indicated. The
201: fiducial massive core in the model of McKee \& Tan (2003) is shown
202: by the dashed line.
203: %Ram pressure from a typical Galactic GMC-GMC
204: % collision ($v=10\:{\rm km\:s^{-1}}$, $n_{\rm H}=500\:{\rm cm^{-3}}$)
205: % is indicated by a horizontal dashed line: note pressures are
206: % converted to an equivalent $\Sigma$ that would create the same mean
207: % pressure in a spherical, self-gravitating cloud, i.e. $P/k=4.3
208: % \times 10^8\Sigma^2\:{\rm K~cm^{-3}}$ (McKee \& Tan 2003).
209: \label{fig:1}}
210: \end{figure}
211: %
212:
213: \section{Star Cluster Formation}
214:
215: \subsection{The Initial Conditions for Star Cluster Formation: Infrared Dark Clouds}
216:
217: We expect the initial conditions for star clusters to be the densest
218: starless gas clouds. Such clouds reveal themselves by absorption of
219: the Galactic diffuse infrared background and have become known as
220: Infrared Dark Clouds (IRDCs) (Egan et al. 1998).
221:
222: One way to measure the physical properties of these clouds is through
223: extinction mapping (Fig.~\ref{fig:irdc}). Assuming the diffuse
224: Galactic infrared emission behind the cloud is similar to that around it
225: and adopting an infrared extinction law and dust to gas
226: ratio (Weingartner \& Draine 2001) allows the measurement of mass
227: surface density, $\Sigma$. A kinematic distance can be
228: measured from $\rm^{13}CO$ line emission (Simon et al. 2001), and thus
229: the physical size and mass of the cloud determined. The cummulative
230: distributions of $M(>\Sigma)$, i.e. the mass that is at surface
231: densities greater than or equal to a particular $\Sigma$, for five
232: typical IRDCs have been measured by Butler, Tan, \& Hernandez, in prep
233: and are shown in Fig.~1. The IRDCs span physical properties
234: similar to those of embedded star clusters (Mueller et al. 2002), although with somewhat lower surface densities, and
235: so are likely to be representative of the initial conditions of star
236: cluster formation.
237:
238: %We can compare the observed distribution of mass as a function of
239: %$\Sigma$ in IRDCs with that in numerical simulations of gas under
240: %different conditions. Fig.~2....
241:
242: \begin{figure}
243: \includegraphics[height=2.65in,width=2.65in,angle=0]{f2a_tan.eps}
244: \includegraphics[height=2.65in,width=2.65in,angle=0]{f2b_tan.eps}
245: %\includegraphics{sigmaband4.eps}
246: \caption{(a) Left: Example IRDC at $l=28.37$, $b=0.07$ and distance
247: 4.9~kpc observed by Spitzer at 8~$\rm \mu m$. Image is $16.5^\prime$ across. (b) Right: $\Sigma$
248: map of the same cloud derived from extinction of the diffuse
249: background (Butler et al., in prep). Intensity scale is in
250: $\rm g\:cm^{-2}$. Note the extinction mapping technique
251: fails where there is a bright source in front of or in
252: the cloud.}
253: \label{fig:irdc}
254: \end{figure}
255:
256: \subsection{The Timescale for Star Cluster Formation}
257:
258: Some models of massive star and star cluster formation involve the
259: global collapse of the protocluster in about one free-fall time
260: (e.g. Bonnell, Vine \& Bate 2004), while other models that include
261: feedback from the forming stars (e.g. Li \& Nakamura 2006) have star
262: formation occuring more gradually over at least several free-fall
263: times. Based on the results of numerical simulations, Krumholz \&
264: McKee (2005) argued that supersonically turbulent gas forms stars at a
265: slow rate of only a few percent of the total gas mass per dynamical or
266: free-fall time. Tan, Krumholz, \& McKee (2006) extended this analysis
267: to spherical clumps and argued that those clumps that
268: eventually turn a high ($\gtrsim30\%$) fraction of their mass into
269: stars must do so over at least several ($\gtrsim 7$) free-fall times.
270:
271: Tan et al. (2006) also summarized the observational evidence in
272: support of slow, quasi-equilibrium star cluster formation: (1) The
273: morphologies of CS gas clumps are round (Shirley et al. 2003); (2) the
274: spatial distributions of stars in embedded, rich, i.e.
275: high-star-formation-efficiency, star clusters show relatively little
276: substructure; (3) the momentum flux from the combined outflows from
277: protostars in forming clusters is relatively small; (4) the age
278: spreads of stars in rich star clusters are much greater than their
279: current free-fall times; (5) in the ONC a dynamical
280: ejection event associated with the cluster has been dated at 2.5~Myr
281: (Hoogerwerf, de Bruijne, \& de Zeeuw 2001), which is much longer than
282: the free-fall time of the present cluster.
283:
284: Note it is the central, high-star-formation-efficiency region of the
285: cluster where we propose that star formation takes place over several
286: to many free-fall times. These regions have short free-fall times (see
287: Fig.~1), $\sim 10^5\:{\rm yr}$. The outskirts of the cluster have much
288: lower densities, longer free-fall times, and star formation here may
289: occur over just one or two free-fall times, as proposed by Elmegreen
290: (2000), before gas is disrupted by feedback from the newly-formed
291: cluster. The global star formation efficiency here will be relatively
292: low and the young stars will exhibit a greater degree
293: of substructure.
294:
295: It has been suggested that the star cluster formation process takes
296: only one or two free-fall times when this time is referenced to the
297: pre-cluster conditions at lower density, and that therefore star
298: cluster formation can be regarded as being the result of dynamic
299: collapse of a cloud and is not a quasi-equilibrium process (Hartmann \&
300: Burkert 2006). This distinction is important because Krumholz, McKee,
301: \& Klein (2005) showed that the process of star formation by
302: competitive accretion cannot be important in virialized, equilibrium
303: clouds. It requires sub-virial conditions associated with global
304: collapse.
305: %(Note, this is only a necessary, but not sufficient,
306: %condition, since stellar feedback, i.e. protostellar outflows may
307: %significantly disrupt the Bondi-Hoyle accretion process.)
308: Several arguments can be made against the global collapse picture: (1)
309: the protocluster gas clouds appear to be approximately virialized
310: (e.g. Shirley et al. 2003); (2) the final distributions of the
311: distances of the stars from the cluster center should reflect the
312: locations at which they formed or be even larger because of gas
313: removal, yet we see newly formed rich star clusters with concentrated,
314: dynamically-relaxed distributions. Huff \& Stahler (2006) found the
315: star formation history of the ONC showed no dependence on the radial
316: distance from the cluster center; (3) again in the ONC, the 2.5~Myr
317: dynamical ejection event (Hoogerwerf et al. 2001) suggests that at
318: this time the cluster was already in a state of high stellar density
319: with a short free-fall time.
320:
321: %Implications of equilibrium star cluster formation.... feedback
322: %confinement, maintenance of turbulence...
323:
324:
325: \subsection{The Initial Cluster Mass Function}
326:
327:
328: The initial cluster mass function (ICMF) is a fundamental property of
329: the star cluster formation process. If there are external triggers,
330: e.g. cloud collisions, supernova blast waves, that initiate star
331: cluster formation, then these may influence the ICMF. It has been
332: suggested that super star cluster formation may be favored in the
333: low-shear environment of dwarf irregulars (Billett, Hunter, \&
334: Elmegreen 2002).
335:
336: To investigate whether the ICMF depends on galactic environment,
337: Dowell, Buckalew, \& Tan (2007) used automated source selection from
338: Sloan Digital Sky Survey (SDSS) data to measured the ICMF at masses
339: $\gtrsim 3\times 10^4\:M_\odot$ in 13 nearby ($\lesssim10$~Mpc) dwarf
340: irregular galaxies, which tend to have relatively low metallicity and
341: shear. Cluster ages, masses and reddening were determined by comparing
342: Starburst99 models with the multi-color photometry. Completeness
343: corrections were made, although these are relatively small for massive
344: young clusters at these distances. Foreground stellar and background
345: galactic contamination were assessed and found to be small. The ICMF
346: was assumed to be equal to the mass function of clusters with ages
347: $\leq 20$~Myr. The same procedure was repeated on SDSS data of several
348: nearby spiral galaxies at similar distances but with higher metallicity
349: and shear. Several hundred clusters were identified from both the
350: dwarf irregular and spiral galaxy samples.
351:
352: The main result is that these samples are stastically
353: indistinguishable from each other, suggesting that the ICMF does not
354: depend on galactic shear or metallicity. We find the ICMF is
355: reasonably well fit by a power law $\frac{dN(M)}{dM} \varpropto
356: M^{-\alpha_M}$ with $\alpha_M \simeq 1.5$ in both dwarf irregular and
357: spiral galaxies. This is somewhat shallower than the power law index
358: of $\alpha_M\simeq2$ that has been found in spiral galaxies by
359: (Larsen 2002) using HST images. This may be due to the lower resolution
360: of the SDSS observations, which lead to blending of clusters that form
361: within $\sim 50$~pc of each other. Nevertheless the similarity of the
362: cluster (or association) mass functions between the galaxy samples
363: suggests that a universal process, perhaps turbulent fragmentation
364: inside GMCs (Elmegreen \& Efremov 1997), is responsible for star
365: cluster formation.
366:
367:
368:
369: \begin{acknowledgments}
370: JCT acknowledges support from CLAS, University of Florida.
371: \end{acknowledgments}
372:
373: \begin{thebibliography}{}
374:
375: \bibitem[]{}
376: {Bally, J., \& Zinnecker, H.} 2005, \textit{MNRAS} 129, 2281
377:
378: \bibitem[]{}
379: {Beuther, H., Churchwell, E.B., McKee, C.F., \& Tan, J.C.} 2006, in \textit{Protostars \& Planets V}, ed. B. Reipurth, Arizona, in press, (astro-ph/0602012)
380:
381: \bibitem[]{}
382: {Beuther, H., \& Schilke, P.} 2004, \textit{Science}, 303, 1167
383:
384: \bibitem[]{}
385: {Beuther, H., Schilke, P., Gueth, F., McCaughrean, M., et al.
386: % Andersen, M., Sridharan, T.K., \& Menten, K.M.
387: } 2002, \textit{A\&A}, 387, 931
388:
389: \bibitem[]{}
390: {Billett, O.H., Hunter, D.A., Elmegreen, B.G.} 2002, \textit{AJ}, 123, 1454
391:
392: \bibitem[]{}
393: {Bonnell, I.A., Bate, M.R., \& Zinnecker, H.} 1998, \textit{MNRAS} 298, 93
394:
395: \bibitem[]{}
396: {Bonnell, I.A., Vine, S.G., \& Bate, M.R.} 2004, \textit{MNRAS} 349, 735
397:
398: \bibitem[]{}
399: {Caselli, P., \& Myers, P.C.} 1995, \textit{ApJ}, 446, 665
400:
401: \bibitem[]{}
402: {Cesaroni, R., Galli, D., Lodato, G., Walmsley, C.M., Zhang, Q.} 2006, in \textit{Protostars \& Planets V}, ed. B. Reipurth, Arizona, in press, (astro-ph/0603093)
403:
404: \bibitem[]{}
405: {Crutcher, R.M.} 2005, in {\it Massive star birth: A crossroads of Astrophysics}, IAU Symp. 227, eds. Cesaroni, Felli, Churchwell, Walmsley, Cambridge: Cambridge University Press, p98
406:
407: \bibitem[]{}
408: {de Wit, W.J., Testi, L., Palla, F., \& Zinnecker, H.} 2005, \textit{A\&A} 437, 247
409:
410: \bibitem[]{}
411: {Dobbs, C.L., Bonnell, I.A., \& Clark, P.C.} 2005, \textit{MNRAS}, 360, 2
412:
413: \bibitem[]{}
414: {Doty, S.D., van Dishoeck, E.F., \& Tan, J.C.} 2006, \textit{A\&A}, 454, L5
415:
416: \bibitem[]{}
417: {Dowell, J.D., Buckalew, B.A., \& Tan, J.C.} 2007, \textit{ApJ}, submitted.
418:
419: \bibitem[]{}
420: {Edgar, R., \& Clarke, C.} 2004, \textit{MNRAS} 349, 678
421:
422: \bibitem[]{}
423: {Egan, M.P., Shipman, R., Price, S., Carey, S., Clark, F., \& Cohen, M.} 1998, \textit{ApJ}, 494, L199
424:
425: \bibitem[]{}
426: {Elmegreen, B.G.} 2000, \textit{ApJ}, 530, 277
427:
428: \bibitem[]{}
429: {Elmegreen, B.G., \& Efremov, Y.} 1997, \textit{ApJ}, 480, 235
430:
431: \bibitem[]{}
432: {Fall, S.M., Chandar, R., \& Whitmore, B.C.} 2005, \textit{ApJ} 631, L133
433:
434: \bibitem[]{}
435: {Garmire, G., Feigelson, E.D., Broos, P. \etal } 2000, \textit{AJ}, 120, 1426
436:
437: \bibitem[]{}
438: {Grosso, N., Feigelson, E.D., Getman, K.V. \etal } 2005, \textit{ApJS}, 160, 530
439:
440: \bibitem[]{}
441: {Hartmann, L., \& Burkert, A.} 2006, \textit{ApJ}, in press, (astro-ph/0609679)
442:
443: \bibitem[]{}
444: {Hoogerwerf, R., de Bruijne, J.H.J., \& de Zeeuw, P.T.} 2001, \textit{A\&A}, 365, 49
445:
446: \bibitem[]{}
447: {Huff, E.M., \& Stahler, S.W.} 2006, \textit{ApJ}, 644, 355
448:
449: \bibitem[]{}
450: {Hunter, T., Brogan, C., Megeath, S., Menten, K., et al.
451: %Beuther, H., Thorwirth,S.
452: } 2006, \textit{ApJ}, 649, 888
453:
454:
455: \bibitem[]{}
456: {Kennicutt, R.C.} 1998, \textit{ApJ} 498, 541
457:
458: \bibitem[]{}
459: {Kratter, K., \& Matzner, C.D.} 2007, \textit{MNRAS}, in press, (astro-ph/0609692)
460:
461: %\bibitem[]{}
462: % {Kroupa, P.} 2002, \textit{Science}m 295, 82
463:
464: \bibitem[]{}
465: {Krumholz, M.R.} 2006, \textit{ApJ} 641, 45
466:
467: \bibitem[]{}
468: {Krumholz, M.R., \& McKee, C.F.} 2005, \textit{ApJ}, 630, 250
469:
470: \bibitem[]{}
471: {Krumholz, M.R., McKee, C.F., \& Klein, R.I} 2005, \textit{Nature}, 438, 332
472:
473: \bibitem[]{}
474: {Krumholz, M.R., McKee, C.F., \& Klein, R.I} 2007, \textit{ApJ}, submitted, (astro-ph/0609798)
475:
476: \bibitem[]{}
477: {Lada, C.J., \& Lada, E.A.} 2003, \textit{Ann. Rev. Astron. \& Astrophys.} 41, 57
478:
479: \bibitem[]{}
480: {Larsen, S.} 2002, \textit{AJ}, 124, 1393
481:
482: \bibitem[]{}
483: {Li, Z-Y., \& Nakamura, F.} 2006, \textit{ApJ}, 640, L187
484:
485: \bibitem[]{}
486: {Massey, P.} 1998, in {\em The Stellar Initial Mass Function}, eds. G. Gilmore, D. Howell, ASP Conf. Ser., 142, ASP: San Francisco, p17
487:
488: \bibitem[]{}
489: {Massi, F., Testi, L., \& Vanzi, L.} 2006, \textit{A\&A} 448, 1007
490:
491: %\bibitem[]{}
492: % {McKee, C.F. \& Tan, J.C.} 2002, \textit{Nature} 416, 59
493:
494: \bibitem[]{}
495: {McKee, C.F. \& Tan, J.C.} 2003, \textit{ApJ} 585, 850
496:
497: \bibitem[{Mueller} et~al., 2002]{mueller2002}
498: {Mueller, K.E., Shirley, Y.L., Evans, N.J., \& Jacobson, H.R.} 2002, \textit{ApJS}, 143, 469
499:
500: \bibitem[]{}
501: {Reid, M.A., \& Wilson, C.D.} 2006, \textit{ApJ}, 644, 990
502:
503: \bibitem[]{}
504: {Shu, F.H., Adams, F.C., \& Lizano, S.} 1987, \textit{Ann. Rev. Astron. \& Astrophys.} 25, 23
505:
506: \bibitem[]{}
507: {Shirley, Y.L., Evans, N.J., Young, K.E., Knez, C., Jaffe, D.T.} 2003, \textit{ApJS}, 149, 375
508:
509: \bibitem[]{}
510: {Simon, R., Jackson, J.M., Clemens, D.P., Bania, T.M., \& Heyer, M.H.} 2001, \textit{ApJ}, 551, 747
511:
512: \bibitem[]{}
513: {Tan, J.C.} 2004, in {\em Star Formation in the Interstellar Medium: In Honor of David Hollenbach, Chris McKee and Frank Shu}, ASP Conf. Proc., 323. eds. D. Johnstone, F.C. Adams, D.N.C. Lin, D.A. Neufeld, and E.C. Ostriker. San Francisco: ASP, p249
514:
515: \bibitem[]{}
516: {Tan, J.C.} 2005, in {\em Massive star birth: A crossroads of Astrophysics}, eds. Cesaroni, R., Felli, M., Churchwell, E., Walmsley, M. (CUP), IAU Symp. 227, 318
517:
518: \bibitem[]{}
519: {Tan, J.C., Krumholz, M.R., \& McKee, C.F.} 2006, \textit{ApJ}, 641, L121
520:
521: \bibitem[]{}
522: {Weingartner, J.C., \& Draine, B.T.} 2001, \textit{ApJ}, 548, 296
523:
524:
525: \end{thebibliography}
526:
527: \begin{discussion}
528:
529: \discuss{Linz}{The speaker mentioned that apparently no current
530: star formation occurs in his sample of IRDCs. Can we be really
531: sure about that, since in most cases, objects are found in IRDCs
532: by Spitzer/MIPS?}
533:
534: \discuss{Tan}{There are two issues here: (1) at each location in the
535: cloud there is a constraint on the embedded luminosity from the lack
536: of flux at $\sim 8\:{\rm \mu m}$, and, without having done detailed
537: calculations, my impression is that for most of the regions of IRDCs
538: in our sample there is no current, active, luminous star formation,
539: i.e. massive star formation, occurring. There could be embedded
540: lower-luminosity sources and it would be useful to probe this
541: population (either with Spitzer/MIPs or with x-rays). (2) IRDCs are
542: not a particularly well-defined class of objects, and there can in
543: fact be bright sources nearby in adjacent clouds or even in part of
544: the same cloud (the cloud in Fig. 2 has such a source). Still, if one
545: were to measure the total light to mass ratios of IRDCs these should on
546: the average be quite low compared to more evolved star-forming clouds.}
547: %I agree that objects can already be embedded in
548: %IRDCs, so I rephrase my claim, so that no cluster made of enhanced
549: %star formation is taking place there right now, and the potential
550: %objects embedded there might be of lower mass.}
551:
552: \discuss{Linz}{Still, these objects embedded there might be lower
553: luminosity now but could develop into high-mass YSOs later on?}
554:
555: \discuss{Tan}{I agree that many or most IRDCs, especially the
556: relatively high column density ones that we are studying, are likely
557: to form star clusters and massive stars in the future.}
558:
559: \discuss{Fukui}{In your turbulent picture, how could you explain the
560: formation of super star clusters?}
561:
562: \discuss{Tan}{Observed super star clusters (SSCs) have $\sim
563: 10^6\:M_\odot$ inside a sphere of radius $\sim 3~{\rm pc}$. One basic open
564: question is whether the initial condition is an essentially starless
565: gas cloud with these properties or whether SSCs form more gradually as
566: smaller clouds (perhaps already forming star clusters) merge with the
567: main cluster. In my opinion, it would be difficult to produce the
568: starless initial condition from typical Galactic GMCs without some
569: kind of synchronized, fast trigger. The escape speed from SSCs is
570: greater than the ionized gas sound speed, so they may be forming with
571: very high efficiency from their parent gas clouds (requiring long
572: formation times [and age spreads] in terms of free-fall times) (Tan \&
573: McKee 2004, in proc. of Cancun Workshop). This longer formation time
574: may allow more time for infall and merger of surrounding gas clouds,
575: and the higher efficiency means less total gas mass is needed to reach
576: the final stellar mass.}
577:
578: %
579: %\discuss{Massey}{I’m wondering if you have considered the expected intrinsic dispersion in absolute
580: %magnitude of WRs --— if you consider the (large) mass range that becomes an
581: %early WN or late WC according to the evolutionary models, wouldn’t you expect a large
582: %dispersion in M$_v$?}
583:
584: %\discuss{van der Hucht}{Indeed, we will be always left with some intrinsic scatter in M$_v$ due
585: %to mass differences within the same spectral subtype. But in my opinion, the current
586: %large dispersion is for a large fraction due to incertainties of the adopted distances of
587: %open clusters and OB associations.}
588: \end{discussion}
589:
590: \end{document}
591: