1: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
2: % Dust-cooling-induced Fragmentation of Low-metallicity Clouds %
3: % by T.Tsuribe & K.Omukai %
4: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
5: \documentclass[apjl]{emulateapj}
6: \usepackage{apjfonts}
7: \slugcomment{To appear in {\it The Astrophysical Journal Letters.}}
8: \lefthead{Tsuribe \& Omukai}
9: \righthead{Fragmentation of Low-metallicity Clouds}
10:
11: \begin{document}
12: %%%%%%%%%%%%%%%%%
13: % TITLE PAGE %%%%
14: %%%%%%%%%%%%%%%%%
15:
16: \title{
17: Dust-cooling--induced Fragmentation of Low-metallicity Clouds
18: }
19:
20: \author{Toru Tsuribe\altaffilmark{1} and Kazuyuki Omukai\altaffilmark{2}}
21:
22: \altaffiltext{1}{Osaka University, Toyonaka, Osaka 540-0053, Japan;
23: tsuribe@vega.ess.sci.osaka-u.ac.jp }
24:
25: \altaffiltext{2}{
26: National Astronomical Observatory of Japan, Mitaka, Tokyo 181-8588, Japan;
27: omukai@yso.mtk.nao.ac.jp }
28:
29: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30: % Abstract & Subject Headings %%%%
31: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
32:
33: \begin{abstract}
34: Dynamical collapse and fragmentation of low-metallicity cloud cores
35: is studied using three-dimensional hydrodynamical calculations, with
36: particular attention devoted whether the cores fragment in the
37: dust-cooling phase or not.
38: The cores become elongated in this phase, being unstable
39: to non-spherical perturbation due to the sudden
40: temperature decrease.
41: In the metallicity range of $10^{-6}-10^{-5}Z_{\odot}$, cores
42: with an initial axis ratio $\ga 2$ reach a critical
43: value of the axis ratio ($\ga 30$) and fragment into
44: multiple small clumps.
45: This provides a possible mechanism to produce low-mass stars
46: in ultra-metal-poor environments.
47: \end{abstract}
48: \keywords{
49: cosmology: theory --- hydrodynamics --- instabilities --- stars: formation}
50:
51: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%
52: % TEXT & Acknowledgements %%%
53: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%
54:
55: %----------------------%
56: \section{Introduction}
57: %----------------------%
58: It has been suggested that the first stars in the universe were very massive,
59: typically $\ga 100M_{\sun}$ (e.g., Abel, Bryan, \& Norman 2000,
60: 2002; Bromm et al. 1999, 2002; Omukai \& Palla 2001, 2003).
61: On the other hand, in present-day local star-forming regions,
62: typical stars are low-mass objects $\la 1M_{\sun}$.
63: One of the major factors distinguishing these two modes of star formation
64: is the degree of metal enrichment.
65: At some point in the history of the universe, metallicity exceeds a
66: critical value and the transition to low-mass star formation mode
67: is thought to take place.
68: Since the exact value of critical metallicity is crucial in modeling
69: such important problems as galaxy formation and
70: reionization of the intergalactic medium etc.,
71: some authors have tried to establish this value.
72: Bromm, Ferrara, Coppi \& Larson (2001) studied the fragmentation of
73: low-metallicity clouds, and concluded that clouds with metallicity
74: $\ga 5\times 10^{-4}Z_{\sun}$ fragment into smaller pieces owing to
75: the fine-structure line cooling of carbon and oxygen.
76: Taking into account possible variations in the elemental abundance
77: ratio, Bromm \& Loeb (2003) derived individual critical abundances
78: of C$^{+}$ and O.
79: Santoro \& Shull (2005) included silicon and iron in a similar analysis.
80: However, due to limitations imposed by inclusion of only the atomic
81: line cooling, these authors only studied the thermal evolution
82: in the low-density regime.
83: The mass scale of such fragments is
84: still high $\ga 10-100M_{\sun}$, and another phase of fragmentation
85: is necessary to produce low-mass ($\la 1M_{\sun}$) objects.
86: Regarding this problem, based on the thermal evolution described
87: by Omukai (2000),
88: Schneider et al. (2002, 2003) pointed out that another episode of
89: fragmentation can be caused by dust cooling at higher densities
90: (say, $\ga 10^{10} {\rm cm^{-3}}$).
91: They concluded that dust-induced fragmentation can take place
92: even with metallicity of $10^{-5\pm 1} Z_{\sun}$ and this produces
93: low-mass ($\la 1M_{\sun}$) fragments.
94: However, their treatment of dynamical evolution was too simple and
95: their prediction of fragmentation remained speculative.
96:
97: In the previous paper (Omukai et al. 2005), using an updated
98: thermal evolution and the linear theory for growth of elongation
99: (Hanawa \& Matsumoto 2000), we evaluated fragmentation mass scale
100: under the assumption that fragmentation takes place when
101: the elongation exceeds a critical value ($\sim 1$) in the linear theory.
102: However, an approximately spherical core, for which the linear theory is
103: applicable, is not subject to fragmentation. Only a highly elongated,
104: filamentary (or disk-like) object is gravitationally unstable and
105: easily fragments (e.g., Larson 1995; Tsuribe \& Inutsuka 1999b).
106: Consequently, a discussion based on linear theory needs to be justified
107: by numerical calculations.
108:
109: In consideration of the above, here we follow the evolution of
110: low-metallicity cores during the dust-cooling phase using three-dimensional
111: hydrodynamical calculations and discuss
112: the conditions for fragmentation into low-mass clumps.
113: Our results confirm that the previous analysis by Omukai et al. (2005)
114: is broadly correct: fragmentation does occur in the dust-cooling phase
115: and sub-solar mass fragments are indeed produced.
116:
117: %--------------------------------------%
118: \section{Numerical Modeling}
119: %--------------------------------------%
120:
121: H$_2$-line cooling induces fragmentation of
122: low-metallicity ($\la 10^{-4}Z_{\sun}$) interstellar clouds
123: around $10^{4} {\rm cm^{-3}}$, and cloud cores are produced.
124: Subsequently, the cores start a so-called run-away collapse,
125: where they contract at almost the free-fall rate.
126: Here we study evolution of such cores using
127: three-dimensional hydrodynamical calculation.
128: Since the cores are already self-gravitating, the dark-matter gravity
129: is not considered.
130: For simplicity, the cores are assumed to be non-rotating and isolated.
131:
132: During the run-away collapse phase, a metal-free pre-stellar core
133: obeys the self-similar flow with $\gamma \simeq 1.1$,
134: as indicated by one-dimensional radiation
135: hydrodynamical calculations (Omukai \& Nishi 1998).
136: Here we choose such a self-similar solution as an unperturbed
137: initial state.
138: The initial central number density is $10^{10}{\rm cm^{-3}}$ where
139: the dust cooling is about to become effective.
140:
141: We add both non-spherical density/velocity and
142: random-velocity perturbations to the initial state.
143: As a non-spherical density perturbation, the sphere is elongated to a
144: prolate with expanding in the $z$-direction and shrinking in $x$- and
145: $y$- directions homologously while the central density is kept constant.
146: We adopt prolate elongation because we find that the initial oblateness
147: leads the amplitude of elongation to saturation in the non-linear
148: regime before fragmentation takes place.
149: Elongation of the core is defined by
150: ${\cal{E}}_0 = a/b - 1$,
151: where $a$ ($b$) is the short (long, respectively) core axis length.
152: Here, cases with an initial elongation of
153: ${\cal{E}}_0 = 0.1, 0.32, 0.56$, and $1.0$ are considered.
154: Non-spherical velocity perturbation is added to the
155: unperturbed flow.
156: The amplitude of non-spherical velocity perturbation is assumed
157: to be proportional to the displacement from the sphere and chosen so
158: that the correct linear growth (or damping) rate is achieved.
159: Additionally, a random velocity field with an amplitude of $0.1c_s$
160: is added, where $c_s$ is the isothermal sound velocity at the center.
161:
162: Solving full thermal equations along with hydrodynamics
163: is numerically too expensive.
164: Here we take the following simplified approach:
165: First, using a one-zone model, we solve thermal and chemical processes
166: under the assumption of free-fall collapse.
167: The model is the same as in Omukai et al. (2005), except the dust model,
168: for which we use dust produced by the pair-instability supernova of a
169: metal-free $195M_{\sun}$ star
170: (Schneider, Salvaterra, \& Ferrara 2004; Schneider et al. 2006).
171: The effective equations of state calculated with this model
172: are shown in Figure 1 for metallicities $10^{-6}, 10^{-5}$,
173: and $10^{-4}Z_{\sun}$.
174: Next, for easier analysis and interpretation of the result,
175: we approximate the above equations of state using
176: the following barotropic relations:
177: the ratio of specific heat
178: $\gamma \equiv d{\rm log} p/d {\rm log} \rho$
179: is given by
180: \begin{equation}
181: \label{eq:eos}
182: \gamma = \left\{
183: \begin{array}{ll}
184: 1.1 & \mbox{$(n < n_{1})$} \\
185: 0.5 & \mbox{$(n_{1} < n< n_{2})$} \\
186: 1.1 & \mbox{$(n_{2} < n < n_{3})$} \\
187: 1.4 & \mbox{$(n_{3} < n)$},
188: \end{array}
189: \right.
190: \end{equation}
191: where ${\log}~n_{1}=6-{\rm log}~Z/Z_{\sun}$,
192: ${\log}~n_{2}=11.1-0.4 {\rm log}~Z/Z_{\sun}$, and
193: ${\log}~n_{3}=8.47-1.2 {\rm log}~Z/Z_{\sun}$.
194: In Figure 1, these relations are also shown by dotted lines.
195: The first phase ($n < n_{1}$) corresponds to the H$_2$-cooling phase,
196: the second ($n_{1}<n<n_{2}$) and third ($n_{2}<n<n_{3}$) to
197: the dust-cooling phase before and after the gas-dust coupling,
198: and the last ($n_{3}<n$) to the adiabatic contraction
199: after the core becomes optically thick at $n_{3}$.
200:
201: %----------------- Figure 1 ---------------------
202: \begin{figure}
203: \plotone{f1.eps}
204: \caption[dummy]{Effective equations of state (EOS) for clouds with
205: $Z/Z_{\sun}=10^{-6}, 10^{-5}$ and $10^{-4}$ (solid).
206: Also shown (dashed) are approximate relations (eq. \ref{eq:eos})
207: used in the hydrodynamical calculation.}
208: \label{fig1}
209: \end{figure}
210:
211: Godunov-type smoothed particle hydrodynamics (SPH) with a second-order
212: Riemann solver are used for hydrodynamics and
213: a Tree method is used for gravitational force
214: (Tsuribe \& Inutsuka 1999a).
215: The number of SPH particles is $N=1.6$ (in Runs A, B, D, F, and G) or
216: $2.3\times 10^{6}$ (in Runs C and E), which is
217: sufficiently large to meet the resolution condition for physical
218: fragmentation (Klein et al 2004).
219:
220: %--------------------------------------%
221: \section{Results}
222: %--------------------------------------%
223: Major parameters in this model are the metallicity $Z$, which
224: determines the equation of state, and the initial amplitude of
225: elongation ${\cal{E}}_0$.
226: The numerical results can be classified into two types
227: according to whether the core fragments or not.
228: In the following, we show the typical models for these cases.
229:
230: %--------------------------------------%
231: \subsection{Cores with Fragmentation}
232: %--------------------------------------%
233: Figure 2 shows the evolution of the core with $Z=10^{-5}Z_{\odot}$ and
234: ${\cal{E}}_0=1.0$.
235: Initially, while the core is close to spherical,
236: the evolution of elongation obeys the prediction by
237: the linear analysis of a $\gamma=1.1$ cloud.
238: For central densities $n \ga n_{1} (=10^{11} {\rm cm}^{-3})$,
239: the temperature decreases rapidly owing to the dust cooling,
240: and the core collapses at almost the free-fall rate.
241: In this phase, the elongation grows in proportion to
242: $\rho^m$ with $m=0.354-0.5$, which is between the rate for spherical
243: pressure-free collapse ($m=0.354$) and that for filament collapse ($m=0.5$).
244: In other words, the rate $m$ is enhanced from the linear value for a
245: sphere owing to a non-linear effect.
246: The equation of state begins to stiffen to $\gamma=1.1$
247: at $n=n_{2} (=10^{13}{\rm cm}^{-3})$,
248: where the dust and gas couple thermally,
249: and become adiabatic ($\gamma=1.4$) at
250: $n=n_{3} (=3\times10^{14}{\rm cm}^{-3})$,
251: where the core becomes optically thick.
252: However, the core continues to elongate owing to inertia until
253: $n \sim 10^{16}{\rm cm}^{-3}$, where the maximum value of elongation
254: ${\cal{E}}_{\rm max}=61$ is attained.
255: A long spindle forms in the high-density core.
256: Around this epoch, the spindle stops collapsing in a cylindrical
257: radial direction and then fragments.
258: Using the scale height of a filament
259: $H=(2 c_s^2/\pi G \bar{\rho_c})^{1/2}$, where $c_s$ and $\bar{\rho_c}$
260: is isothermal sound speed and density averaged over the $z$-axis,
261: the interval between each fragment is well described by $2 \pi H$,
262: corresponding to maximum growing mode in the linear stability
263: analysis of equilibrium filaments
264: (Nagasawa 1987; Inutsuka \& Miyama 1997).
265: The filament fragments into four pieces.
266: Since the fragments have an infall velocity along the $z$-axis, they merge
267: with each other at the center, owing to small initial random velocities.
268: This is partly because we do not include effects of rotation and
269: angular momentum.
270: In spite of the merging of fragments at the center,
271: the number of fragments does not decrease, at least
272: within the present calculation,
273: since new fragments continue to be born on the
274: outskirts.
275: %
276: We also studied a case with the same parameters ($Z, {\cal{E}}_0$),
277: but with a different seed for the random number.
278: In this case, positions of density peaks were altered, but
279: the distance between each fragment remained almost the same.
280: The dynamical evolution was also similar.
281:
282: %------------------- Figure 2 --------------------
283: \begin{figure*}
284: %\epsscale{0.65} % without apjfonts
285: \epsscale{0.75} % with apjfonts
286: %\plotone{f2.eps}
287: \plotone{f2small.eps}
288: \caption[dummy]{
289: Density distribution in the $x$-$y$ plane on the midplane
290: (top panels), $x$-$z$ plane of $y=0$ (bottom panels)
291: for the case with ${\cal{E}}_0=1.0$ and
292: $Z=10^{-5}Z_{\odot}$ at six different stages.
293: The color scale denotes the density in the logarithmic scale.
294: The maximum number density and time are shown in each panel.
295: }\label{fig2}
296: \end{figure*}
297: In the case of lower metallicity
298: ($Z=10^{-6}Z_{\odot}$, ${\cal{E}}_0=1.0$),
299: the maximum elongation becomes smaller (${\cal{E}}_{\rm max} \sim 40$)
300: because of a weaker effect of dust cooling.
301: Fragmentation is marginally observed.
302: There are two fragments.
303:
304: %--------------------------------------%
305: \subsection{Cores without Fragmentation}
306: %--------------------------------------%
307: Models with smaller initial elongation exhibit different evolution.
308: As an example, we show one model with $Z=10^{-5}Z_{\odot}$ and
309: ${\cal{E}}_0=0.32$ in Figure 3.
310: In the initial stage ($\gamma=1.1$), the evolution of elongation is
311: very similar to examples with fragmentation (\S 3.1).
312: Subsequently, in the cooling phase
313: ($n \ga n_{1}=10^{11} {\rm cm}^{-3}$), the elongation grows.
314: The growth rate, however, remains close to the linear growth rate
315: ($m=0.354$) because of the small initial elongation.
316: Then, the growth of elongation slows down at
317: $n_{3} \simeq 3\times 10^{14} {\rm cm}^{-3}$,
318: where the equation of state becomes adiabatic.
319: The maximum elongation reaches $14$ in this case, but this is not
320: enough to cause fragmentation.
321: Thereafter, the core becomes more spherical with contraction,
322: instead of becoming cylindrical.
323: We also followed the evolution in the subsequent accretion stage
324: by concentrating on the central region.
325: No fragmentation was observed in our calculation.
326:
327: %------------------ Figure 3 -------------------------
328: \begin{figure*}
329: %\epsscale{0.65} % without apjfonts
330: \epsscale{0.75} % with apjfonts
331: %\plotone{f3.eps}
332: \plotone{f3small.eps}
333: \caption[dummy]{
334: Same as Fig. 1 but for the case with ${\cal{E}}_0=0.32$ and
335: $Z=10^{-5}Z_{\odot}$.
336: }\label{fig3}
337: \end{figure*}
338:
339: In this calculation, the maximum value of axis ratio reaches 14.
340: This exceeds both the critical value for fragmentation $\pi$ and
341: maximum growing mode $2\pi$ for fragmentation in the linear analysis
342: for uniform infinite filaments.
343: Despite this, the filament does not fragment because of
344: the central density concentration in our case.
345: Accretion onto the central density peak is faster than fragmentation
346: on the outskirts.
347: In the cases with ${\cal{E}}_0=0.1$, no fragmentation is observed
348: for any of the metallicities.
349:
350: %------------------------------------%
351: \section{Conclusion and Discussion}
352: %------------------------------------%
353: We have studied the fragmentation process of low-metallicity
354: ($10^{-6}-10^{-4}Z_{\sun}$) cloud cores at high densities
355: ($\ga 10^{10}{\rm cm^{-3}}$).
356: Cores with initial axis ratios greater than about two become
357: elongated enough to fragment via dust-cooling, even
358: with metallicities of $10^{-6}-10^{-5}Z_{\sun}$.
359: This provides a possible mechanism to produce low-mass stars
360: in ultra-metal-deficient environments.
361: %
362: The results of our calculations are summarized in Table 1.
363: As a rule of thumb, fragmentation takes place if the core elongation
364: reaches a critical value ${\cal{E}}_{\rm crit} \ga 30$.
365: According to linear analysis for a filament of infinite length, namely,
366: without density gradient along the long axis, an equilibrium filament is
367: gravitationally unstable and subject to fragmentation if the axis ratio
368: is greater than $\pi$ (Nagasawa 1987).
369: In our case, the filament is of finite length.
370: The density gradient toward the center along the long axis makes
371: the critical value for fragmentation several times larger than that for
372: a filament of infinite length as discussed in \S 3.2.
373: In Table 1, the maximum values of elongation in the linear analysis
374: ${\cal{E}}_{\rm max, L}$ are also presented.
375: We see that if ${\cal{E}}_{\rm max, L}$ exceeds $\sim 3$,
376: the elongation of the core will eventually reach
377: ${\cal{E}}_{\rm crit} \ga 30$ and the core fragments.
378: It should be noted that the actual maximum value of elongation and that
379: in linear theory are not reached at the same time because of non-linear
380: effects.
381: The linear maximum value of elongation
382: ${\cal{E}}_{\rm max, L}$ is attained at the end of the rapid
383: dust-cooling phase (i.e., $n=n_{2}$), since the elongation
384: grows (or decays) if $\gamma < 1.09$ ($\gamma > 1.09$, respectively),
385: in the linear theory (Hanawa \& Matsumoto 2000; Lai 2000).
386: On the other hand, the actual maximum is reached slightly after the core
387: becomes adiabatic (i.e., $n \ga n_{3}$).
388:
389: Although we have demonstrated by detailed numerical work
390: that dust-induced fragmentation indeed occurs under some circumstances,
391: some points still remain to be improved.
392: To avoid too large a dynamic range,
393: we were forced to take the initial state of our calculation
394: at a rather high density of $10^{10}{\rm cm^{-3}}$ and
395: parameterize the elongation of cores at this point.
396: The question of how frequently such an initial condition is realized
397: remains uncertain.
398: The evolution after fragmentation is also to be studied.
399: The final mass of stars is determined as a result of complex dynamical
400: evolution such as merging and accretion.
401: In this phase, angular momentum would play an important role,
402: although we have neglected effects of rotation for simplicity.
403: For example, a disk may fragment into multiples during
404: the accretion phase (e.g., Saigo, Matsumoto, \& Umemura 2004).
405: These issues should be investigated in a future work.
406:
407: \acknowledgments
408: Numerical computations were carried out on VPP5000 at the Astronomical
409: Data Analysis Center of the National Astronomical Observatory of Japan
410: and in the computational facilities at Osaka University.
411: This research was supported in part by Grants-in-Aid for Young
412: Scientists (B) 14740129 (TT) by the Ministry of Education, Culture,
413: Sports, Science, and Technology of Japan (MEXT).
414:
415: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
416: %%%%%%% References %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
417: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
418:
419: \begin{thebibliography}{}
420: \bibitem[]{407} Abel, T., Bryan, G.~L., \& Norman, M.~L.\ 2002, Science, 295, 93
421: \bibitem[]{408} Bromm, V., Coppi, P. S., \& Larson, R. B. 2002, \apj, 564, 23
422: \bibitem[]{409} Bromm, V., Ferrara, A., Coppi, P. S., \& Larson, R. B. 1999, \mnras, 328, 969
423: \bibitem[]{411} Hanawa, T., \& Matsumoto, T. 2000, \pasj, 52, 241
424: \bibitem[]{412} Klein, R. I., Fisher, R. \& McKee, C. F. 2004, RMxAC, 22, 3
425: \bibitem[]{413} Lai, D. 2000, \apj, 540, 946
426: \bibitem[]{414} Larson, R. B. 1985, \mnras, 214, 379
427: \bibitem[]{415} Nagasawa, 1987, Prog. Theor. Phys., 77, 635
428: \bibitem[Omukai(2000)]{Omukai00} Omukai, K.\ 2000, \apj, 534, 809
429: \bibitem[Omukai \& Nishi(1998)]{ON98} Omukai, K. \& Nishi, R. \ 1998, \apj, 518, 64
430: \bibitem[Omukai et al.(2005)]{OTSF05} Omukai, K., Tsuribe, T., Schneider, R. \& Ferrara, A. 2005, \apj, 626, 627
431: \bibitem[]{420} Saigo, K., Matsumoto, T., \& Umemura, M. 2004, \apj, 615, L65
432: \bibitem[]{421} Santoro, F., \& Shull, J. M. 2005, \apj, submitted (astro-ph/0509101)
433: \bibitem[]{423} Schneider, R., Ferrara, A., Natarajan, P., \& Omukai, K. 2002, \apj, 571, 30
434: \bibitem[]{424} Schneider, R., Ferrara, A., Salvaterra, R., Omukai, K, \& Bromm, V. 2003, \nat, 422, 869
435: \bibitem[]{425} Schneider, R., Salvaterra, R., \& Ferrara, A. 2004, \mnras, 351, 1379
436: \bibitem[]{427} Schneider, R., Inoue, A. K., Omukai, K. \& Ferrara, A. 2006, in preparation
437: \bibitem[Tsuribe \& Inutsuka(1999)]{Tsuribe99a} Tsuribe, T.~\& Inutsuka, S.\ 1999, \apj, 523, L155
438: \bibitem[Tsuribe \& Inutsuka(1999)]{Tsuribe99b} Tsuribe, T.~\& Inutsuka, S.\ 1999, \apj, 526, 307
439:
440: \end{thebibliography}
441:
442: %\end{document}
443:
444: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
445: %%%%%% Table %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
446: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
447: %\input{tab1.tex}
448: %
449: %\begin{deluxetable}{cccccccc}
450: \begin{deluxetable*}{cccccccc}
451: \tablecaption{Model Parameters and Results
452: \label{tab:model} }
453: \tablehead{
454: \colhead{Run} &
455: \colhead{$\log{Z/Z_{\odot}}$} &
456: \colhead{${\cal{E}}_0$} &
457: \colhead{${\cal{E}}_{\rm max}$} &
458: \colhead{$\log n_{\rm max}$} &
459: \colhead{${\cal{E}}_{\rm max, L}$} &
460: \colhead{Fragmentation} &
461: }
462: \startdata
463: A & -6 & 0.56 & 16 & 16.8 & 1.5 & No \\
464: B & -6 & 1.0 & 30 & 16.6 & 2.7 & Yes \\
465: C & -5 & 0.32 & 14 & 16.4 & 1.6 & No \\
466: D & -5 & 0.56 & 27 & 16.1 & 2.8 & Yes \\
467: E & -5 & 1.0 & 61 & 16.0 & 4.9 & Yes \\
468: F & -4 & 0.32 & 34 & 15.6 & 2.9 & Yes \\
469: G & -4 & 0.56 & 56 & 15.3 & 5.1 & Yes \\
470: \enddata
471: \tablenotetext{~}{NOTE: Model parameters are metallicity
472: $\log{Z/Z_{\odot}}$ and the initial elongation ${\cal{E}}_0$.
473: ${\cal{E}}_{\rm max}$ is the maximum value of elongation, which is
474: attained at the number density $n_{\rm max}$. ${\cal{E}}_{\rm max, L}$
475: is the maximum value of elongation in the linear theory. }
476: %\end{deluxetable}
477: \end{deluxetable*}
478: %
479:
480: \end{document}
481: