1: \documentclass[letter]{aa}

2: %\documentclass[referee]{aa}

3: %\documentstyle[12pt]{article}

4: %\begin{document}

5: 

6: %\documentclass[12pt]{article}

7: %\usepackage{francais}

8: %\usepackage{epsfig}

9: \usepackage{color} % Para usar colores en el texto

10: 

11: \usepackage{ulem} % for underlining and cross outs

12: 

13: \usepackage{graphicx}

14: \usepackage{amsmath,epsfig,rotating,amssymb}

15: 

16: \begin{document}

17: %\input{psfig.tex}

18: 

19: \def \cha{\widehat}

20: \def \div{{\rm div}}

21: \def \grad{{\rm {\bf grad} }}

22: \def \kms{{\rm ~km~s}^{-1}}

23: \def \pcc{{\rm ~cm}^{-3}}

24: \def \psc{{\rm ~cm}^{-2}}

25: \def \pr{{\it permanent}  regime }

26: \def \rot{{\rm {\bf rot} }}

27: \def \VS{{V\'azquez-Semadeni}}

28: %\textwidth 6.in

29: %\oddsidemargin 0.in

30: %\evensidemargin 0.in

31: %\topmargin 0.in

32: %\headheight 0.5in

33: %\headsep 0.2in

34: %\textheight 9.in

35: %\parindent 0.2in

36: %\pagestyle{myheadings}

37: 

38: %\input{psfig.tex}

39: 

40: \author{Hennebelle P. \inst{1}, Banerjee R. \inst{2},

41: V\'azquez-Semadeni E. \inst{3}, Klessen R. \inst{2}, Audit E. \inst{4}}

42: 

43: \institute{Laboratoire de radioastronomie, UMR 8112 du

44: CNRS, 

45: \newline {\'E}cole normale sup{\'e}rieure et Observatoire de Paris,

46:  24 rue Lhomond, 75231 Paris cedex 05, France 

47: \and Zentrum fuer Astronomie der Universitat Heidelberg, Institut 

48: fuer Theoretische Astrophysik, 69120 Heidelberg, Germany 

49:  \and Centro de Radioastronom\' \i a y Astrof\'\i sica, Universidad Aut\'onoma de M\'exico, Apdo Postal 3-72 Morelia, 58089, M\'exico

50: \and  Service d'Astrophysique, CEA/DSM/DAPNIA/SAp, Centre d'\'Etudes de Saclay,

51: l'Orme les Merisiers, 91191 Gif-sur Yvette Cedex, France}

52: 

53: 

54: \offprints{  P. Hennebelle \\

55: {\it patrick.hennebelle@ens.fr}   }

56: 

57: 

58: \title{From the warm magnetized atomic medium to molecular clouds}

59: 

60: \authorrunning{Hennebelle et al.}

61: 

62: %\runningtitle{Formation of molecular clouds}

63: 

64: 

65: \abstract

66: { It has  recently been proposed that giant molecular

67: complexes form at the sites where streams of diffuse warm

68: atomic gas collide at transonic velocities.}

69: {We study the  global statistics of molecular clouds formed by large scale

70:    colliding    flows

71: of warm neutral atomic interstellar  gas under

72: ideal MHD conditions.  The flows deliver material as well

73: as  kinetic energy and trigger thermal instability leading eventually to 

74: gravitational collapse.}

75: {We perform adaptive  mesh refinement MHD simulations which,  for

76:       the first time in this context, treat self-consistently cooling and self-gravity.}

77: {The clouds formed in the simulations  develop a highly

78: inhomogeneous density and temperature structure, with cold dense

79: filaments and clumps

80: condensing from converging flows of warm atomic gas. In the clouds, the

81: column density probability density distribution (PDF) peaks at $\sim 2

82: \times 10^{21} \psc$ and decays rapidly at higher

83: values; the magnetic intensity correlates weakly with density from $n

84: \sim 0.1$ to $10^4 \pcc$, and then varies roughly as $n^{1/2}$ for higher

85: densities. }

86: {The global statistical properties of such molecular

87:     clouds are reasonably consistent with

88: observational determinations.

89:  Our numerical simulations suggest that molecular clouds

90: formed by the moderately supersonic collision of warm atomic gas streams.}

91: \keywords{Magnetohydrodynamics  (MHD) --   Instabilities  --  Interstellar  medium:

92: kinematics and dynamics -- structure -- clouds} 

93: 

94: \maketitle

95: 

96: 

97: \section{Introduction}

98: The formation of molecular clouds is one of the key steps for the star

99: formation process.  A large number of studies investigate the

100: internal dynamics of molecular clouds (e.g. see the review by MacLow

101: \& Klessen 2004), but only a few investigations address the problem of

102: their formation itself  (e.g. see the review by

103: Hennebelle, Mac Low \& V\'azquez-Semadeni  2008). This is partly due to the difficulty in

104: treating the large range of spatial scales relevant in this problem

105: and partly due to uncertainties on the mechanisms at the origin of

106: their formation.  During the last decade, the idea has emerged that

107: the molecular clouds may be formed at the onset of a large scale

108: converging flow of atomic gas (e.g. Ballesteros-Paredes et al. 1999)

109: with the active role of thermal instability (Hennebelle \& P\'erault

110: 1999, Koyama \& Inutsuka 2000, 2002, Audit \& Hennebelle 2005, Heitsch

111: et al. 2005, V\'azquez-Semadeni et al. 2006).  Indeed, since the

112: density of molecular clouds is much larger than the mean ISM density,

113:  large scale flows are a viable explanation to excite density

114: enhancements.

115: The origin  

116: of these flows is however unclear and  may be not unique.

117:  Most likely they  arise from turbulent fluctuations 

118: or gravitational instability occuring at  large scales. 

119: 

120: Recently large multi-dimensional   non-magnetic

121:  numerical simulations   have been

122: performed (Hennebelle \& Audit 2007, V\'azquez-Semadeni et al. 2007, 

123: Heitsch et al. 2008, the last two including self-gravity)

124:  with the aim of studying in  detail the formation of 

125: dense gas from a large flow of warm neutral medium,

126: resolving down to the star forming scales.  

127: 

128: 

129: In this letter, we present 

130: the first results of large adaptive mesh refinement (AMR), MHD simulations

131: performed with the codes RAMSES (Teyssier 2002, Fromang et al. 2006) 

132: and FLASH (Fryxell et al. 2000). 

133: These   are the first simulations which, starting from the WNM, 

134: include  magnetic

135: field, cooling, self-gravity, and thanks to the AMR scheme,  have 

136: sufficient spatial resolution  to resolve individual, high

137:   density, clouds.  

138: The molecular clouds we observe in these simulations are

139:   self-consistently generated by thermal instabilities, turbulence 

140: and  gravitational contraction. 

141: We stress that  

142: the internal structure and the

143:   turbulent properties of these molecular clouds 

144: are not the result of  an ad hoc assumption on the external

145:   turbulent driving.  By performing these 

146: numerical simulations, we expect to tackle  unsolved  and outstanding 

147: questions such  as what drives turbulence in molecular

148: clouds; what is the gas density and temperature distribution, 

149: and what is the structure of the magnetic field in these objects{\bf ?}

150: In this letter we  report on the most important global

151: properties of the clouds. In subsequent articles (Banerjee et al. 2008),  

152: we will give a more

153: detailed analysis of the cloud formation, structure and evolution,

154: and on the efficiency of star formation in the simulations.

155: 

156:  

157: In section 2, we describe the numerical set up and the initial conditions

158: whereas in section 3 we present our results and  preliminary 

159: comparisons with observations. Section 4 concludes  this paper.

160: 

161: 

162: \begin{figure}

163: \includegraphics[width=6.5cm]{hist_tot.ps}

164: \caption{Top and second panels: Column density and density PDF 

165: in the simulation. 

166: Third, fourth and fifth panels: Temperature, 

167: magnetic intensity and magnetic intensity variance as a function of 

168: gas density.  The {\it solid, dotted, dashed, and

169: dash-dotted} lines show the distributions at times $t=3.7$, 7.7,

170: 12.0 and 13.8 Myr, respectively. The straight line  in

171: the fourth panel 

172: shows a distribution proportional to  $n^{1/2}$. }

173: \label{histo}

174: \end{figure}

175: 

176: %

177: 

178: 

179: \section{Numerical setup and initial conditions}

180: The numerical  simulation presented in this letter  has

181: been performed with  

182:  the AMR code RAMSES using the HLL solver. Ramses is a second order

183: Godunov scheme and uses the constraint transport method to ensure

184:   ${\rm div} B =0$ (see Fromang et al. 2006).

185: Starting with an initial resolution of 256$^3$ cells, 2 levels of 

186: refinement are allowed during the calculation leading to an effective 

187: 1024$^3$ numerical simulation. The criterion used to refine the grid is a 

188: simple density threshold of 50 cm$^{-3}$ for the first level and 200 cm$^{-3}$

189: for the second one. This ensures that the dense gas is uniformly 

190: resolved.  With the box size being about 50 pc, this leads to 

191:  a spatial resolution of about 0.05 pc. The total number of cells

192: in the simulation is about $\simeq 4 \times 10^7$. About 25,000

193: timesteps have been performed for a total of 30,000 cpu hours. 

194: 

195: 

196: To mimic a large scale converging 

197: flow (e.g. Audit \& Hennebelle 2005),

198: a converging velocity field is imposed at the left and right faces 

199: of the simulation box, on top of which velocity modulations have been 

200: superimposed. The velocity of each  incoming flow is  twice

201: the sound speed of the WNM, leading to a total velocity  difference of

202: about 40 km s$^{-1}$ within the box.

203: The amplitude of the modulation is about  a factor of two, and  it is

204:   periodic with a spatial frequency of 10 pc.

205: The boundary conditions are periodic for the 4 remaining faces.

206: Initially, the density and temperature are respectively 

207: 1 cm$^{-3}$ and about 

208: 8000 K, which  are also the values imposed at  the left and right faces.

209: The velocity is initially equal to zero through the box.

210: The magnetic field is uniform initially and parallel to the x-axis, therefore

211: aligned with the incoming velocity field and has an intensity of about

212:  5 $\mu$G, corresponding to equipartition between magnetic and thermal pressure

213: initially.

214: The cooling is due to atomic species as described in Audit \& Hennebelle 

215: (2005). Molecular cooling and H$_2$ formation are not treated at this stage.

216: As we will see, this nevertheless leads to  reasonable temperature and 

217: density  distributions. 

218: 

219: \setlength{\unitlength}{1cm}

220: \begin{figure*}

221: \begin{picture}(0,13.5)

222: \put(0,6.5){\includegraphics[width=7cm]{ray_xy_00020_1.ps}}

223: \put(0,0){\includegraphics[width=7cm]{dens_vit_xy_00020_1.ps}}

224: \put(9,6.5){\includegraphics[width=7cm]{mag_xy_00020_1.ps}}

225: \put(9,0){\includegraphics[width=7cm]{Temp_xy_00020_1.ps}}

226: \end{picture}

227: \caption{Top left panel: column density. Top right panel:  Magnetic intensity 

228: and its xy-components (indicated as arrows) in the $z=0$ plane. Bottom left panel: density

229:   and velocity fields in the $z=0$ plane. Bottom right panel: temperature

230:  in the $z=0$ plane.}

231: \label{field}

232: \end{figure*}

233: 

234: 

235: 

236: %\begin{figure}

237: %\includegraphics[width=7cm]{ray_xy_00020.ps}

238: %\caption{  Column density.}

239: %\label{ray}

240: %\end{figure}

241: 

242: 

243: %\begin{figure}

244: %\includegraphics[width=8cm]{3D_2.ps}

245: %\caption{3D view of the density field.}

246: %\label{3Dview}

247: %\end{figure}

248: 

249: 

250: %\begin{figure}

251: %\includegraphics[width=7cm]{dens_vit_xy_00020.ps}

252: %\caption{Density and velocity fields in the $z=0$ plane.}

253: %\label{dens_vit}

254: %\end{figure}

255: 

256: %\begin{figure}

257: %\includegraphics[width=7cm]{Temp_xy_00020.ps}

258: %\caption{Temperature field in the $z=0$ plane.}

259: %\label{temperature}

260: %\end{figure}

261: 

262: 

263: %\begin{figure}

264: %\includegraphics[width=7cm]{mag_xy_00020.ps}

265: %\caption{Magnetic intensity and its xy-components in the $z=0$ plane.}

266: %\label{magnetic}

267: %\end{figure}

268: 

269: 

270: 

271: \section{Results}

272: Figure~\ref{histo} shows the column density and density pdf as 

273: well as temperature, magnetic intensity and its variance as a function 

274: of density at time 3.71, 7.7, 12 and 13.79 Myr. Because of the mass 

275: injection, the total mass increases continuously within the simulation box 

276: from about 3000 $M_s$ initially to roughly 10 times this value at 

277: time 13.79 Myr. At time 3.71 and 7.7 Myr, the largest density reached in 

278: the simulation is  between a few times 10$^3$ cm$^{-3}$

279:  and $10^4 \pcc$, whereas at time 

280: 13.79 Myr gravity has taken over and triggers gravitational 

281: collapse producing much higher densities. This indicates that the

282: cloud  should start

283: forming stars roughly 12 Myr after the collision of the converging flow 

284: has occurred ($t\simeq$ 1 Myr in the simulation).

285: At later times, the cloud keeps forming stars while the total mass continues

286: to increase. This is qualitatively in good agreement with the observations

287: reported by Blitz et al. (2007)  for the LMC, that the

288: masses of GMCs with little star formation activity are smaller than

289: those of GMCs with strong activity.

290: %regarding the mass of GMC in the  Magellanic 

291: %clouds being  smaller for the GMC{{\color{blue}s that}}  have little 

292: %star formation than for

293: %the one which actively form stars. 

294: We note that the

295: duration of  7 Myr, they infer for the first  massive starless phase, is similar to the 

296: timescale of our simulation roughly estimated between time 7.7 Myr and 13.79

297: Myr. We stress, however, that precisely defining the

298: ``birth'' time of the molecular cloud in our simulation is an elusive

299: task since the mass is an  increasing function of time. 

300: In future papers we will address

301: this point from an observational perspective. 

302: 

303: The  mass weighted PDF of the column density distribution peaks at about $2 \times 10^{21}$ cm$^{-2}$

304: and drops rapidly for higher values. It is interesting to note that

305: this is similar to what has recently been  inferred by Goldsmith et al. 

306: (2008) for the Taurus molecular cloud (see their figure 8).

307: 

308: The temperature drops  rapidly for densities between 

309: 3 and 30 cm$^{-3}$ where it reaches a value of about 50 K. It then slowly 

310: decreases  down to about 10 K for densities higher than 10$^4$ cm$^{-3}$. 

311: Note that since UV shielding and molecular cooling are not considered here, 

312:  the temperature in the dense gas  is

313: probably overestimated. This would imply  

314: that the average density of the cold clumps should probably be a little

315: higher.  Explicitly treating

316: H$_2$ formation would have the same effect (Glover \& MacLow 2007).

317: 

318: For densities smaller than $\simeq 1000$ cm$^{-3}$, the magnetic 

319: intensity increases very smoothly  with density whereas 

320: for density larger than $\simeq 1000$ cm$^{-3}$, it is roughly 

321: proportional to $\sqrt{\rho}$.  Indeed,  the lower density gas is

322:   magnetically sub-critical and mainly flows along the magnetic field

323:   lines without compressing magnetic flux. On the other hand,  the 

324:  high density gas is supercritical and the magnetic flux is compressed along

325: with the dense gas under the influence of the gravitational force.

326: Future studies will have to investigate whether ambipolar diffusion modifies 

327: this behaviour significantly.

328:  The variance of the magnetic intensity increases 

329: smoothly  with density and ranges from about 

330: one third to half the mean magnetic intensity. We note that this 

331: magnetic intensity distribution appears to be very similar to what 

332: has been inferred from observations (Troland \& Heiles 1986, Crutcher 1999)

333: 

334: 

335: Figure~\ref{field} shows 

336: the  column density along the z-axis, the density and velocity field at 

337: $z=0$, and the temperature 

338: field in the same plane at time $t=7.7$ Myr after the beginning of 

339: the simulation. The column density  reveals that 

340: the cloud has a complex internal structure made of filaments and dense

341: clumps of density  between 100-1000 cm$^{-3}$,

342:  embedded  in a more diffuse phase. This is even more clearly 

343: apparent in the density  and temperature cuts which show 

344: that the clumps are relatively isolated and embedded into the warm diffuse

345: phase.

346: % which, together with the ram pressure exerted by the converging flows,

347: % helps confining the clumps (see Banerjee et al. 2008). 

348: This suggests that the molecular clouds are not

349: homogeneous isothermal media,

350: but rather, consist of a complex array of

351: filaments and clumps, with denser gas being colder, and therefore

352: with little or no excess of thermal pressure over the surrounding gas. 

353: %This

354: %fact makes the dense structures more long-lived than in an isothermal

355: %medium, with their dynamics depending on the fluctuations of the turbulent ram

356: %pressure. 

357: %an ensemble of reasonably well

358: %defined dense clumps (although 

359: %numerical resolution is probably largely insufficient to provide an accurate

360: % description of the interface between warm and cold gas). 

361: 

362: Interestingly, the density of the 

363: warm gas embedded in the molecular cloud

364:  is higher than  in the outer medium ($n \simeq 1 $ cm$^{-3}$)

365:  and can be as large as 3-4 cm$^{-3}$. 

366: Indeed,  this gas has been previously shocked 

367:  and it is in the process of cooling as it moves towards

368: the dense cold regions (\VS\ et al. 2006; Hennebelle \& Audit 2007).

369: %to the ram pressure which compresses it.    

370: From a comparison between the column density and temperature distribution 

371: in the $z=0$ plane, 

372: we note that the warm gas is deeply embedded in the molecular cloud.  

373: Note that in this work, the UV field is assumed to be constant.

374:  Although this is obviously not a good assumption for the 

375: dense gas, we see that since the filling factor of the cloud appears to be 

376: small, this is certainly a fair assumption  for the WNM even when it is 

377: deeply embedded inside the cloud. Moreover, 

378:  the higher temperatures are sometimes found at the edge of the clumps 

379: at the onset of the accretion shocks which occurs when the WNM flow 

380: encounters a dense clump. This clearly indicates that 

381: the dissipation of mechanical energy plays an active role in the heating 

382: of the warm phase. Altogether, this is in good agreement with the 

383: picture proposed by Hennebelle \& Inutsuka (2006) except that the 

384: mechanical energy which heats the warm phase is the kinetic energy of

385: the shocks rather than the energy of the MHD waves (possibly underestimated 

386: in this work since the ion-neutral drift is not treated). 

387: Note that the finding of interclump medium being low density atomic hydrogen

388: ($n<4-10$ cm$^{-3}$) is consistent with the estimate of Williams et al. (1995) 

389:  for the Rosette molecular cloud.

390: 

391: Figure~\ref{field} also shows the magnetic intensity in the $z=0$ plane

392:  as well 

393: as its xy-components. In the external medium the magnetic field remains

394: much more uniform than in the dense regions, where its direction 

395: fluctuates significantly. 

396: Since the field was initially

397: uniform, this implies that the turbulent

398: motions are able to significantly distort the field. 

399: 

400: 

401: \section{Conclusion}

402: 

403: We have presented the results of AMR MHD simulations aiming to

404: describe self-consistently the formation of a molecular cloud from a

405: converging flow of warm diffuse atomic hydrogen.  Our simulations

406: suggest a complex multi-phase structure of the molecular clouds which

407: consists  of cold and dense clumps embedded into a warm atomic phase.

408: The simulations reproduce reasonably well the observed variations of magnetic

409:   intensity with density and column density distribution.

410: % As shown in the companion letter, they also

411: %reproduce well various clump properties. 

412: Finally, we suggest that star

413: formation may start in the cloud while it is still accreting

414: material. This would imply that the mass of GMCs vary with

415: time, in

416: good agreement with recent observations (Blitz et al. 2007), and with the

417: results of previous non-magnetic simulations of the phenomenon.

418: 

419: 

420: \begin{thebibliography}{99}

421: 

422: \bibitem{audit-hennebelle}

423: Audit, E., \& Hennebelle, P. 

424: {2005, A\&A 433, 1}

425: %

426: 

427: 

428: %\bibitem{de avillez}

429: %de Avillez, M., \& Breitschwerdt, D.

430: %{\ 2005, A\&A 436, 585}

431: %

432: 

433: \bibitem{BVS99}

434: Ballesteros-Paredes, J., V\'azquez-Semadeni, E., \& Scalo, J. 

435: {\ 1999a, ApJ, 515, 286}

436: %

437: 

438: %\bibitem{ballesteros99}

439: %Ballesteros-Paredes, J., Hartmann, L., \& V\'azquez-Semadeni, E. 

440: %{\ 1999b, ApJ, 527, 285}

441: %

442: 

443: \bibitem{robi}

444: Banerjee, R., V\'azquez-Semadeni, E., Hennebelle, P., Klessen, R., 

445: {\ 2008, A\&A, {\rm in preparation}}

446: 

447: 

448: \bibitem{Blitz_etal07}

449: Blitz, L., Fukui, Y., Kawamura, A., Leroy, A., Mizuno, N., \&

450: Rosolowsky, E. 2007, in Protostars and Planets V., eds. B. Reipurth,

451: D. Jewitt, \& K. Keil (Tucson: U. of Arizona Press) 81.

452: %

453: 

454: 

455: %\bibitem{Bourke_etal01} 

456: %Bourke, T. L., Myers, P. C., Robinson, G., \& Hyland, A. R. 

457: %{\ 2001, ApJ 554, 916}

458: %

459: 

460: \bibitem{crutcher99}

461: Crutcher, R., 

462: {\ 1999, ApJ, 520, 706}

463: 

464: %\bibitem{Crutcher04}

465: %Crutcher, R. M.

466: %{\ 2004, Ap\&SS, 292, 225}

467: %

468: 

469: \bibitem{fromang}

470: Fromang, S., Hennebelle, P., Teyssier, R.,  

471: {\ 2006, A\&A, 457, 371}

472: 

473: 

474: 

475: 

476: \bibitem{glover07b} Glover, S. C. O. G., \& Mac Low, M.-M. 2007b, ApJ,

477:   659, 1317

478: %

479: 

480: 

481: \bibitem{goldsmith}

482: Goldsmith, P., Heyer, M., Narayanan, G., Snell, R., Li, D., Brunt, C., 

483: {\ 2008, arXiv:0802.2206}

484: 

485: 

486: %\bibitem{hartmann01}

487: %Hartmann, J., Ballesteros-Paredes, J., \& Bergin, E. 

488: %{\ 2001, ApJ, 562, 852}

489: %

490: \bibitem[\protect\citeauthoryear{Hennebelle, Mac Low, \&

491: \VS}{2008}]{HMV08} Hennebelle, P., Mac Low, M.-M., \& \VS\ 2007, in

492: Structure Formation in the Universe: Galaxies, Stars, Planets,

493: ed. G. Chabrier (Cambridge: Cambridge University Press), in press

494: (arXiv:0711.2417)

495: 

496: 

497: %\bibitem{heiles_troland05}

498: %Heiles, C., \& Troland, T.

499: %  {\ 2005, ApJ, 624, 773}

500: %

501: 

502: 

503: \bibitem{heitsch05}

504: Heitsch, F., Burkert, A., Hartmann, L., Slyz, A., \& Devriendt, J.

505: {\ 2005, ApJ 633, 113}

506: %

507: 

508: \bibitem{heitsch08}

509: Heitsch, F., Hartmann, L., Slyz, A., Devriendt, J.,  \& Burkert, A.

510: {\ 2008, ApJ 674, 316 }

511: %

512: 

513: \bibitem{hen-aud}

514: Hennebelle, P., \& Audit, E.

515: {\ 2007, A\&A, 465, 431}

516: %

517: 

518: 

519: \bibitem{hen1}

520: Hennebelle, P., \& P\'erault, M.

521: {\ 1999, A\&A 351, 309}

522: %

523: 

524: %\bibitem{hen2}

525: %Hennebelle, P., \& P\'erault, M.

526: %{\ 2000, A\&A 359, 1124}

527: %

528: 

529: \bibitem{hen-inu06}

530: Hennebelle, P., Inutsuka, S.-I.

531: {\ 2006, ApJ, 647, 404}

532: 

533: 

534: %\bibitem{joung}

535: %Joung, M. K. R., \& Mac Low, M.-M. 2006, ApJ, 653, 1266

536: %

537: 

538: 

539: \bibitem{koyoma1}

540: Koyama, H.,  Inutsuka, S.

541: {\ 2000, ApJ 532, 980}

542: %

543: 

544: \bibitem{koyoma2}

545: Koyama, H.,  Inutsuka, S.

546: {\ 2002, ApJ 564, L97}

547: %

548: 

549: %\bibitem{LG03}

550: %Li, D.,  Goldsmith, P. F.

551: %{\ 2003, ApJ 585, 823}

552: %

553: 

554: 

555: \bibitem{mk04} Mac Low, M.-M.,  Klessen, R. S. 2004, Rev.\ Mod.\

556:   Phys., 76, 125

557: %

558: 

559: 

560: %\bibitem{PN99}

561: %Padoan, P., Nordlund, \AA.

562: %{\ 1999, ApJ, 526, 279}

563: %

564: 

565: \bibitem{Fryxell}

566: Fryxell, B., Olson, K., Ricker, P., et al.

567: {\ 2000, ApJS, 131, 273}

568: 

569: 

570: \bibitem{teyssier}

571: Teyssier, R., 

572: {\ 2002, A\&A, 385, 337}

573: 

574: 

575: \bibitem{troland}

576: Troland, T., \& Heiles, C.

577: {\ 1986, ApJ, 301, 339}

578: %

579: 

580: 

581: \bibitem{vs07}

582: V\'azquez-Semadeni, E., G\'omez, G. C., Jappsen, A. K.,

583: Ballesteros-Paredes, J., Gonz\'alez, R. F., \& Klessen, R. S. 2007,

584: ApJ, 657, 870

585: %

586: 

587: \bibitem{vazquez-semadeni}

588: V\'azquez-Semadeni, E., Ryu, D., Passot, T., Gonz\'alez, R., \& Gazol, A.

589: {\ 2006, ApJ, 643, 245}

590: %

591: 

592: \bibitem{williams}

593: Williams, J., Blitz, L., \& Stark, A.

594: {\ 1995, ApJ, 451, 252}

595: %

596: 

597: 

598: \end{thebibliography}

599: 

600: 

601: \end{document}

602: 

603: 

604: 

605: 

606: 

607: %

608: % Ancienne version

609: %

610: 

611: 

612: 

613: 

614: 

615: 

616: 

617: 

618: 

619: 

620: 

621: 

622: 

623: