1: \documentclass[12pt,onecolumn]{pasj00}
2: %\draft
3: %\documentclass[12pt,preprint]{aastex}
4: %\usepackage{emulateapj5,epsfig}
5:
6: %\citestyle{aa}
7: \usepackage{graphicx}
8: \newcommand{\Vec}[1]{\mbox{\boldmath $#1$}}
9:
10: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
11: %%%%%% (1) Title Page %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
12: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
13:
14: \SetRunningHead{Maki \& Susa}{Dissipation of Magnetic Flux in Primordial
15: Star Formation}
16:
17: \title{Dissipation of Magnetic Flux in Primordial Star Formation:\\
18: From Run-away Phase to Mass Accretion Phase}
19: \author{Hideki Maki}
20: \affil{Department of Physics, Rikkyo University,
21: Nishi-Ikebukuro, Tokyo 171-8501}
22: \email{hide-mk@jcom.home.ne.jp}
23:
24: \author{Hajime Susa}
25: \affil{Department of Physics, Konan University,
26: Okamoto, Kobe 658-8501}
27: \email{susa@center.konan-u.ac.jp}
28:
29: \KeyWords{accretion, accretion disks --- diffusion --- early universe ---
30: stars: formation --- stars: magnetic fields}
31: \Received{$\langle$reception date$\rangle$}
32: \Accepted{$\langle$acception date$\rangle$}
33: \Published{$\langle$publication date$\rangle$}
34:
35: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
36: %%%%%% (2) Abstract & Subject Heading %%%%%%%%%%%%%%%%%%%%%%%%%%%
37: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
38: \begin{document}
39: \maketitle
40: \begin{abstract}
41: We investigate the dissipation of magnetic flux
42: in primordial star-forming clouds throughout their collapse
43: including the run-away collapse phase
44: as well as the accretion phase.
45: We solve the energy equation and
46: the non-equilibrium chemical reactions in the
47: collapsing gas, in order to obtain the radial distribution of the
48: ionized fraction during the collapse.
49: As a result, we find the ionized fraction is high enough for the
50: magnetic field to couple with the gas throughout the evolution of the
51: cloud.
52: This result suggests that the jet formation from protostars as well
53: as the activation of magneto-rotational instability in the accretion disk
54: are enabled in the presence of the cosmological seed magnetic flux
55: proposed by \citet{langer03}.
56: \end{abstract}
57:
58:
59: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
60: %%%%%% (3) Text & Acknowledgement %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
61: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
62:
63: \section{INTRODUCTION}
64: \label{intro}
65: Typical mass or initial mass function of population III stars are
66: fundamental parameters that have great impacts on subsequent structure
67: formation of the universe. Those stars are expected to be as massive as
68: $100-1000\MO $,
69: %which could have significant contribution to the
70: %source of the reionization \citep{Sokasian04}.
71: which ionize/dissociate the surrounding media. As a result, star formation
72: activities in the neighbourhood of the stars are highly regulated
73: (e.g. \citet{Susa07} and the references are therein).
74:
75: There are two chief reasons that the population III stars are expected
76: to be very massive. First one comes from numerical
77: studies by several authors
78: \citep{nakamura99,bromm99,abel00,nakamura01,bromm02}.
79: Those studies reveal that the prestellar cores formed as fragments of
80: primordial gas are as massive as $\sim 10^3 \MO-10^4 \MO$.
81: %, which is much largerthan $1\MO$.
82: Second ground is the very high accretion rates of those
83: stars, which is as high as $10^{-3}-10^{-2}\MO{\rm yr}^{-1}$. These two
84: facts are the direct consequence of relatively high temperature
85: ($\sim 1000$K) of primordial gas because of inefficient cooling by
86: H$_2$.
87:
88: On the other hand,
89: abundance pattern of the hyper metal-poor stars seems to be more
90: consistent with that of ``faint'' supernovae as remnants of less
91: massive ($\sim 25 \MO$) population III
92: stars\citep{Christlieb04,Frebel05,Iwamoto05}.
93: Recent theoretical studies
94: suggest the possibilities for the formation of such less massive
95: population III stars.
96: If the primordial gas was once ionized, enhanced fraction of H$_2$
97: causes more efficient cooling and HD formation. Since HD molecules have
98: lower excitation energy than that of H$_2$, gas can be cooled below
99: 100K. As a result, the mass of the fragments could be smaller than
100: $100\MO$\citep{oshea05,JB06}. \citet{oshea07} also demonstrates that such low
101: mass population III stars cloud be formed directly from the cosmological
102: density fluctuations.
103:
104: Those new ideas are quite interesting and promising,
105: however, we still do not know the
106: actual mass of population III stars formed from $\sim 10^3 \MO -10^4 \MO$
107: prestellar cores, which are commonly found in cosmological simulations.
108: Since they are only $10^4-10^6 {\rm cm^{-3}}$, we have
109: to follow the subsequent evolution of the collapsing cloud.
110: Further evolution of prestellar cores is first investigated by
111: \citet{omukai98}, and they found that the collapse proceeds in a run-away
112: fashion and converges to Larson-Penston type similarity solution
113: \citep{Larson69,Penston69,suto88} with polytropic index
114: $\gamma\simeq 1.09$.
115: They also find the mass accretion rate is very
116: large compared to the present-day forming stars, although spherical
117: symmetry is assumed in their radiation hydrodynamic simulations. Recently,
118: \citet{Yoshida06} have performed cosmological simulations in which the
119: run-away collapsing core is traced up to a very dense regime ($n_{\rm H}
120: \gg 10^{10}{\rm cm^{-3}}$), taking the Sobolev type line transfer
121: approximations. They also find basically consistent results in 3D
122: cosmological simulations with previous 1D results by \citet{omukai98}.
123:
124: However, most of the mass of protostar is accumulated in the accretion
125: phase, which has not been traced especially in multi-dimensional
126: simulations. Recent numerical simulations performed by \citet{saigo05}
127: show that disks would be formed at the center of collapsing
128: primordial clouds,
129: which might result in the disks surrounding the protostars, or
130: binaries. Thus, the actual mass of a population III star should depends on the
131: mechanism of the angular momentum transport in the accretion disk.
132: There are a few possibilities of the angular momentum transport mechanism
133: such as gravitational torque by the nonaxisymmetric structures in the
134: accretion disk, the interaction among the
135: fragments of the disk \citep{stone00,bodenheimer00},
136: and the turbulent viscosity
137: triggered by Magneto-Rotational Instability (MRI)
138: \citep{hawley92,sano98,sano01}.
139: As for MRI induced turbulence, the strength of magnetic field in
140: the disk is the key quantity to activate the instability\citep{TM04,TB04}.
141: It is also worth noting that
142: recent MHD simulations by \citet{machida06} suggests the
143: possibility of bipolar jets from proto-population III stars, which also
144: could suppress the mass accretion onto the central core. The formation
145: of jets also requires the presence of some level of magnetic field.
146: Therefore, it is quite important to assess the magnetic field
147: strength brought into the accretion disk from initial weak cosmological
148: seed field (e.g. \citet{langer03}).
149:
150: \citet{MS04} investigated this issue by solving detailed chemical
151: reaction rate equations coupled with energy equation in run-away collapsing
152: core. They found that magnetic field is always frozen to the collapsing
153: core in case the strength of initial magnetic field is smaller than
154: $10^{-5}(n_{\rm H}/10^3~{\rm cm^{-3}})^{0.55}~{\rm G}$. This is
155: comparable to the maximal strength of magnetic field which allow the
156: clouds to collapse. Therefore they conclude magnetic field is always
157: frozen to the {\it collapsing} cloud in the run-away collapsing core.
158: They also evaluate the minimal strength of magnetic field which is
159: required to activate MRI assuming magnetic field is frozen to the gas
160: not only in the run-away phase, but also in the accretion phase.
161:
162: This assumption is based upon the argument that the temperature of the
163: accretion flow would rise faster than that of the core, because of the
164: shock heating. As a result, the ionization degree is expected to be
165: higher than those in run-away phase, which guarantee the gas to be frozen
166: to the magnetic field lines.
167: However, such heating is only important in the final
168: phase of accretion, where the flow settle onto the accretion disk.
169: In order to investigate the coupling of matter and magnetic field in the
170: accretion phase, we need to calculate the actual thermal and chemical
171: evolution of accretion flow.
172: Aside from the issue on ionization degree,
173: the ambipolar diffusion velocity is proportional to the square
174: inverse of density ($v_{\rm B}^{\rm amb}\propto \rho^{-2}$).
175: As a result,
176: the diffusion velocity increases rapidly in the mass accretion phase,
177: since the density at a fixed radius decreases as the accretion proceeds.
178: Thus, we need a detailed treatment to test the coupling between the gas
179: and magnetic field in accretion phase.
180:
181: In this paper, we investigate the dissipation of magnetic field in
182: collapsing primordial gas cloud from run-away phase to accretion phase,
183: by solving detailed thermal and chemical rate equations.
184: In the next section, we describe the formulations employed.
185: In \S\ref{results}, results of our calculations are shown. The
186: formation of jets and activation of MRI is discussed
187: in \S\ref{discussion}. Final section is devoted to summary.
188:
189:
190: %\section{SIMPLISTIC ANALYSIS WITH SELF-SIMILARITY SOLUTIONS}
191: %First of calculating the energy equation and the chemical reaction,
192: %we assess briefly the dissipation of the magnetic flux
193: %in the primordial gas clouds at the mass accretion phase
194: %using similarity solutions.
195:
196: %A polytrope what is collapsing by the self-gravity
197: %with polytropic index $\gamma$ is obedient to similarity solutions.
198: %The similarity solutions including the mass accretion phase
199: %is investigated by \citet{suto88},
200: %and two differential equations is derived.
201: %In general, those equations are unable to be solved
202: %and must be integrated numerically.
203: %Here, we aim to get an analytical understanding of
204: %the behavior of the dissipation of the field easily,
205: %and therefore pay attention to the asymptotic
206: %similarity solutions at the central vicinity
207: %and the far-off distance from center.
208:
209: %\subsection{Asymptotic Similarity Solutions at Accretion Phase}
210: %In this subsection, we assume that the EOS of the polytrope is
211: %$p(r,t)=K\rho^\gamma(r,t)$, where $K$ is constant
212: %with respect to a time $t$,
213: %that is zero at the time when the central density is infinity
214: %at the run-away phase and the accretion phase just start.
215:
216: %First, we consider the asymptotic limit $r\to\infty$ at same time
217: %to view the behavior far from the center.
218: %According to \citet{suto88}, in this case,
219: %the distributions of a density and a velocity are given by
220: %\begin{eqnarray}
221: %\rho(r,t) &=& \frac{Ak^{1/(2-\gamma)}}{4\pi G}r^{-2/(2-\gamma)}, \\
222: %u(r,t) &=& B\sqrt{k}\left(\frac{r}{\sqrt{k}}\right)^{(1-\gamma)/(2-\gamma)},
223: %\end{eqnarray}
224: %where $A$ and $B$ are constants of integration,
225: %and $k$ is the value defined by $K\equiv k(4\pi G)^{\gamma-1}$.
226: %From these equations, we can see that the dependence of the density
227: %on radius $r$ is $\rho\propto r^{-2/(2-\gamma)}$,
228: %and that of the velocity is $u\propto r^{(1-\gamma)(2-\gamma)}$.
229: %Next, we consider the asymptotic solution at $r\to0$.
230: %In this case, the density and velocity are taken by
231: %\begin{eqnarray}
232: %\rho(r,t) &=&
233: % \frac{r^{-3/2}M_{\rm c}(t)^{(2-3\gamma)/(8-6\gamma)}}{4\pi G}
234: % \sqrt{\frac{(4-3\gamma)m(0)k^{3/2}}{2}} \nonumber \\
235: % & & \qquad\qquad\qquad\times
236: % \left[\frac{(4-3\gamma)G}{k^{3/2}m(0)}\right]^{(2-3\gamma)/(8-6\gamma)},
237: %\end{eqnarray}
238: %\begin{equation}
239: %u(r,t) = -\sqrt{2\frac{GM_{\rm c}(t)}{r}},
240: %\end{equation}
241: %where $M_{\rm c}(t)$ is the central mass defined by
242: %\begin{equation}
243: %M_{\rm c}(t)\equiv M(r=0,t) = \frac{k^{3/2}t^{4-3\gamma}}{G(4-3\gamma)}m(0),
244: %\end{equation}
245: %and $m(0)$ is the nondimensional value of the central mass at $r=0$.
246: %Hence, we see that the density and the velocity have relation to
247: %the radius and the center mass(time):
248: %$\rho\propto r^{-3/2}M_{\rm c}(t)^{(2-3\gamma)/(8-6\gamma)}$,
249: %and $u\propto r^{-1/2}M_{\rm c}(t)^{1/2}$, respectively.
250:
251: %\subsection{Limiting Behavior of Dissipation Using Similarity Solutions}
252: %\label{sec:simple argument}
253: %In the next, we compare the drift velocities of the ambipolar diffusion
254: %and the ohmic loss with the above gas velocity.
255: %If the ion fraction is $y_{\rm e} = y_{\rm i}$,
256: %the dependences of the drift velocities are
257: %$v_{{\rm B}x}^{\rm amb}
258: %\propto y_{\rm e}^{-1}y_{\rm n}^{-1}n_{\rm H}^{-2}B^2r^{-1}$,
259: %$v_{{\rm B}x}^{\rm ohm}\propto y_{\rm e}^{-1}y_{\rm n}r^{-1}$, respectively,
260: %where $y_{\rm n}$ is the fraction of neutral particles
261: %(cf. Nakano \& Umebayashi 1986).
262: %Therefor, the ratios of the drift velocities to the falling velocity of
263: %the gas are
264: %\begin{eqnarray}
265: %\frac{v_{{\rm B}x}^{\rm amb}}{u(r,t)}
266: % &\propto& y_{\rm e}^{-1}y_{\rm n}^{-1}B^2r^{(2\gamma+1)/(2-\gamma)},\\
267: %\frac{v_{{\rm B}x}^{\rm ohm}}{u(r,t)}
268: % &\propto& y_{\rm e}^{-1}y_{\rm n}r^{(2\gamma-3)/(2-\gamma)},
269: %\end{eqnarray}
270: %at the accreting envelope($r\to\infty$),
271: %\begin{eqnarray}
272: %\frac{v_{{\rm B}x}^{\rm amb}}{u(r,t)}
273: % &\propto& y_{\rm e}^{-1}y_{\rm n}^{-1}B^2r^{5/2}
274: % M_{\rm c}^{(9\gamma-8)/(8-6\gamma)},
275: %\label{eq:simple drift amb}\\
276: %\frac{v_{{\rm B}x}^{\rm ohm}}{u(r,t)}
277: % &\propto& y_{\rm e}^{-1}y_{\rm n}r^{-1/2}M_{\rm c}^{-1/2}.
278: %\label{eq:simple drift ohm}
279: %\end{eqnarray}
280: %Taking the $\gamma=1.09$, the behavior at $r\to\infty$ is that
281: %$v_{\rm B}$ is smaller than the gas velocity as the radius $r$ is larger.
282: %Then, if the dissipation could be neglected at the center,
283: %the magnetic field is tend to not dissipate at the envelope,
284: %except the dependence on the electric fraction.
285: %At $r\to0$, comparing at the same radius,
286: %the ohmic loss is tend to decrease as the central mass
287: %grows with the progress of the accretion.
288: %Against that, we can see that
289: %the ambipolar diffusion is tend to increase
290: %if $8/9<\gamma<3/2$.
291: %Because the specific heat of the primordial gas
292: %accreting to the center will be $8/9<\gamma<3/2$,
293: %it seems that the ambipolar diffusion may operate effectively
294: %as the central mass is increasing by the gas accretion.
295: %The dissipation rate, however, is strongly dependent on
296: %the electric fraction.
297: %Hence, it is not clear that the frozen condition
298: %is also valid at the accretion phase,
299: %and then we must calculate the motion of the accreting gas
300: %to be included the heating/cooling processes
301: %and the non-equilibrium chemical reactions.
302:
303:
304: \section{METHOD OF CALCULATIONS}
305: \label{models}
306: In order to evaluate the coupling of the gas and the magnetic field, we need
307: to assess the amount of ions and electrons in collapsing primordial
308: cloud. The collapse of the cloud is expected to proceed in run-away
309: fashion (run-away phase) in the beginning, followed by the mass
310: accretion phase after the formation of rotationally supported disk at
311: the center. In this paper, we follow the chemical and thermal evolution
312: of the materials to form star, throughout the two phases. We assume
313: spherically symmetric collapse of the progenitor gas, although we expect
314: the formation of rotationally supported disk in the very dense regime
315: ($n_{\rm H} \gg 10^{10} {\rm cm^{-3}}$). It is also assumed that the
316: magnetic field is so weak that the dynamics of the collapsing gas
317: is not affected by the magnetic force.
318:
319: \subsection{Run-away Collapse Phase}
320: \label{runaway}
321: The run-away collapse phase is traced by solving 1-dimensional hydrodynamics.
322: The equation of continuity is
323: \begin{equation}
324: \frac{dm}{dr}=4\pi r^2\rho,
325: \end{equation}
326: whereas the equation of motion is
327: \begin{equation}
328: \frac{Du}{Dt}=-4\pi r^2\frac{\partial p}{\partial m}-\frac{Gm}{r^2},
329: \end{equation}
330: where $m$ is the mass within radius $r$, $\rho$ is the density of the cloud,
331: $u$ is the velocity and $p$ is the pressure.
332: Since the equation of state of collapsing primordial gas in run-away phase
333: is known to be approximated by polytrope with $\gamma=1.09$ \citep{omukai98}, we employ following equation:
334: \begin{equation}
335: p=K\rho^\gamma, \quad \gamma=1.09
336: \end{equation}
337: We solve above set of equations by spherically symmetric Lagrangian
338: hydrodynamics code developed by ourselves, following the Piecewise
339: Parabolic Method (PPM) described in \citet{CW84}.
340:
341: \subsection{Mass Accretion Phase}
342: \label{accretionphase}
343: The dynamics of accretion phase could be approximated by simple free-fall,
344: since the flow velocity is supersonic.
345: Basic equations of free-falling accretion flow consists of
346: the equation of continuity and the equation of motion:
347: \begin{equation}
348: 4\pi\rho(t_0,r_0)r_0^2dr_0=4\pi\rho(t,r)r^2(t)dr,
349: \label{eq:continuity at accretion phase}
350: \end{equation}
351: \begin{equation}
352: \frac{d^2r\left(t\right)}{dt^2} = -\frac{GM_0}{r^2(t)},
353: \label{eq:eq. of motion in accretion phase}
354: \end{equation}
355: where $r$, $t$ and $t_0$ represent
356: the position of fluid element, time, and the time when the mass
357: accretion starts, respectively.
358: $M_0$ denotes the mass within $r_0$ at $t=t_0$:
359: \begin{equation}
360: M_0 = \int_0^{r_0}4\pi r'_0\rho_0dr'_0.
361: \end{equation}
362: The initial radius $r_0$ is defined as $r_0 \equiv r(t_0)$.
363: Therefore, the the solution of above equation of motion is given as
364: $r=r(t;t_0,r_0)$.
365: The velocity of the fluid element is given by
366: \begin{equation}
367: u(t;t_0,r_0) = -\sqrt{2\mathcal{E}_0+2\frac{GM_0}{r(t)}},
368: \end{equation}
369: where $\mathcal{E}_0$ is the total energy defined as
370: $\mathcal{E}_0\equiv u^2(t_0)/2-GM_0/r_0$.
371: Using equation (\ref{eq:continuity at accretion phase}),
372: the density at $(t,r)$ is given as
373: \begin{equation}
374: \rho(t,r) = \frac{r_0^2}{r^2}
375: \frac{\rho_0}{\left(\partial r/\partial r_0\right)_t}\label{eq:density in accretion phase}.
376: \end{equation}
377: $(\partial r/\partial r_0)_t$ is the partial differentiation by $r_0$
378: keeping $t$ fixed. An explicit expression of this term is
379: given in appendix \ref{appendix_a}.
380: In order to clarify the validity of the free-fall approximation, it is compared with the similarity solution
381: with $\gamma=1.09$ \citep{suto88} in accretion phase. The initial
382: condition of the accretion flow is set as the final state of the
383: run-away collapse with $\gamma=1.09$. Figure \ref{fig:free-fall v.s. polytrope} illustrates the density and velocity distributions of several
384: snapshots. It is obvious that the free-fall approximation can describe
385: the matter distribution in accretion phase very well for polytropic gas.
386: In other words, the thermal
387: evolution has little effects on the dynamics of the accretion flow. Thus,
388: the dynamics can be approximated by free-fall formula, irrespective of
389: the internal energy equation.
390:
391: \begin{figure}[tbh]
392: \begin{center}
393: \FigureFile(13cm,10cm){fig1.eps}
394: %\includegraphics[width=0.7\linewidth]{fig1.eps}
395: \caption{
396: Time evolution of density/velocity profiles in accretion
397: flow are shown. Dotted lines and marks represent the result with free-fall
398: approximation,
399: whereas the solid lines denote the results from the similarity
400: solution in accretion phase with $\gamma=1.09$.
401: }
402: \label{fig:free-fall v.s. polytrope}
403: \end{center}
404: \end{figure}
405:
406: In the accretion flow, the effective polytropic index is basically
407: unknown.
408: Thus, we have to solve the following energy equation for each fluid
409: element:
410: \begin{equation}
411: \frac{d\varepsilon}{dt} =
412: -p\frac{d}{dt}\left(\frac{1}{\rho}\right)-\mathcal{L}^{\rm(net)}.
413: \end{equation}
414: In this expression, $p$ denotes the pressure, which is related to the internal energy
415: $\varepsilon$ by the equation of state:
416: \begin{equation}
417: p = (\gamma_{\rm ad}-1)\rho\varepsilon,
418: \end{equation}
419: where $\gamma_{\rm ad}$ is the adiabatic exponent, $\mathcal{L}^{\rm
420: (net)}$ denotes the net energy loss rate per unit mass.
421:
422: We take into account the rovibrational line cooling by $\rm H_2$($\mathcal{L}_{\rm line}$),
423: the continuum radiation from the gas ($\mathcal{L}_{\rm cont}$),
424: and cooling and heating associated with chemical reactions
425: ($\mathcal{L}_{\rm diss},~\mathcal{G}_{\rm H^-},~\mathcal{G}_{\rm
426: H^+_2}$ and $\mathcal{G}_{\rm 3body}$). $\mathcal{L}^{\rm (net)}$ is the
427: sum of all these contributions.
428:
429: The cooling rate by the rovibrational transition of hydrogen molecules
430: are assessed by the fitting formula given by \citet{galli98} in case the
431: lines are optically thin. In the optically thick regime, we employ the
432: escape probability formalism, as described in equations
433: (12)-(15) in \citet{omukai00}, except that we use actual velocity gradient
434: in our calculations.
435:
436:
437: %Here the continum opacity $\tau_{\rm cont}$ is evaluated as
438: %\begin{eqnarray}
439: %\tau_{\rm cont}= \int^\infty_r\kappa_{\rm gas}\rho(t,r) dr,
440: %\end{eqnarray}
441: %where $\kappa_{\rm gas}$ is the Planck mean opacity of gas \citep{lenzuni91}, which includes bound-free absorption,
442: %free-free absorption, photodissociation,
443: %Rayleigh scattering, and collision-induced absorption.
444:
445: The cooling rate owing to the continuum radiation is given by
446: %\begin{equation}
447: %\mathcal{L}_{\rm cont} = -\frac{4\sigma}{\mu m_{\rm p}\kappa_{\rm
448: % gas}\rho } \frac{1}{r^2}\frac{\partial}{\partial
449: % r}\left(r^2\lambda\frac{\partial T^4}{\partial r}\right)
450: %\end{equation}
451: %where $\lambda$ denotes the flux limiter introduced by \citet{LP81},
452: \begin{equation}
453: \mathcal{L}_{\rm cont} = 4\sigma \kappa_{\rm gas} T^4
454: \end{equation}
455: where $\sigma$ denotes the Stefan-Boltzmann constant,
456: $\kappa_{\rm gas}$ is the Planck mean opacity of gas \citep{lenzuni91}, which includes bound-free absorption,
457: free-free absorption, photodissociation,
458: Rayleigh scattering, and collision-induced absorption.
459: This formula is valid in case the accreting gas is optically thin for
460: continuum radiation. We confirmed that the optical depth is smaller than
461: unity throughout the accretion phase.
462:
463: Cooling due to the latent heat of H$_2$ dissociation is given by
464: \begin{equation}
465: \mathcal{L}_{\rm diss}=4.48\frac{n_{\rm H}}{\rho}
466: \left(\frac{dy_{\rm H_2}}{dt}\right)_- {\rm eV\;s^{-1}\;g^{-1}},
467: \end{equation}
468: where $n_{\rm H}$ is the number density of hydrogen nuclei,
469: and $(dy_{\rm H_2}/dt)_-$ is the dissociation rate of $\rm H_2$.
470: %In the same way as that, an ionization absorb
471: %the same energy as the binding energy:
472: %that is, $13.6\rm\;eV$ for H ionization,
473: %$24.6\rm\;eV$ for He ionization,
474: %$79.0\rm\;eV$ for $\rm He^+$ ionization,
475: %and $5.39\rm\;eV$ for Li ionization.
476:
477: On the other hand, the gas is heated when
478: a hydrogen molecule is formed.
479: The heating rate per unit mass is given by
480: \begin{eqnarray}
481: & \mathcal{G}_{\rm H^-} = \frac{3.53}{\rho}
482: \left(\frac{dy_{\rm H_2}}{dt}\right)_{\rm H^-}
483: \frac{n_{\rm H}}{1+n_{\rm cr}/n_{\rm H}}{\rm\;eV\;s^{-1}\;g^{-1}}, \\
484: & \mathcal{G}_{\rm H^+_2} = \frac{1.83}{\rho}
485: \left(\frac{dy_{\rm H_2}}{dt}\right)_{\rm H^+_2}
486: \frac{n_{\rm H}}{1+n_{\rm cr}/n_{\rm H}}{\rm\;eV\;s^{-1}\;g^{-1}}, \\
487: & \mathcal{G}_{\rm 3body} = \frac{4.48}{\rho}
488: \left(\frac{dy_{\rm H_2}}{dt}\right)_{\rm 3body}
489: \frac{n_{\rm H}}{1+n_{\rm cr}/n_{\rm H}}{\rm\;eV\;s^{-1}\;g^{-1}},
490: \end{eqnarray}
491: where $(dy_{\rm H_2}/dt)_{\rm H^-}$, $(dy_{\rm H_2}/dt)_{\rm H^+_2}$
492: and $(dy_{\rm H_2}/dt)_{\rm 3body}$ are
493: the formation rates of $\rm H_2$ by H$^-$ process, H$_2^+$ process and
494: three body reactions \citep{HM79}.
495: $n_{\rm cr}$ is the critical density defined as
496: \begin{eqnarray}
497: n_{\rm cr} = \frac{10^6}{T^{1/2}}
498: \Biggl\{1.6y_{\rm H}\exp\left[-\left(\frac{400}{T}\right)^2\right]
499: + 1.4y_{\rm H}\exp\left[-\left(\frac{1.2\times10^4}{T+1200}\right)\right]
500: \Biggr\}^{-1}\;{\rm cm^{-3}}.
501: \end{eqnarray}
502: \subsection{Chemical Reactions}
503: Since the dissipation of the magnetic flux,
504: as well as the temperature of the gas,
505: strongly depends on the chemical abundances,
506: we have to solve the non-equilibrium chemical reaction rate equations
507: coupled with equation of motion and energy equation described in previous
508: subsections \ref{runaway} and \ref{accretionphase}.
509:
510:
511: The evolution of the fraction of species $i$ is followed by
512: solving the equations,
513: \begin{eqnarray}
514: \frac{dy_i}{dt} &=& \sum_{l=1}^{24}\sum_{m=1}^{24}n_{\rm H}k_{lm}y_ly_m
515: + \sum_{l=1}^{24}\sum_{m=1}^{24}\sum_{n=1}^{24}n_{\rm H}k_{lmn}y_ly_my_n,
516: \nonumber \\
517: & &\qquad\qquad\qquad\qquad\qquad\quad(i=1,2,3,\cdots,24),
518: \end{eqnarray}
519: where $y_i\equiv n_i/n_{\rm H}$ is the fraction of species $i$,
520: $k_{lm}\;{\rm [cm^3\;s^{-1}]}$ and
521: $k_{lmn}\;{\rm [cm^6\;s^{-1}]}$ are
522: the reaction rate coefficients
523: with respect to two-body processes and three-body processes, respectively.
524: In our calculation, we include 24 species:
525: $\rm e^-$, $\rm H^+$, H, $\rm H^-$, $\rm H_2$, $\rm H_2^+$,
526: $\rm H_3^+$, D, $\rm D^+$, $\rm D^-$, HD, $\rm HD^+$, $\rm H_2D^+$,
527: He, $\rm He^+$, $\rm He^{++}$, $\rm HeH^+$, Li, $\rm Li^+$,
528: $\rm Li^{++}$, $\rm Li^{3+}$, $\rm Li^-$, LiH, and $\rm LiH^+$.
529: We employ the latest reaction rate coefficients in the
530: following papers, \citet{galli98}, \citet{omukai00},
531: \citet{stancil98}, \citet{flower02} and \citet{lepp02}.
532: As for the radiative recombination, we use the rate coefficients based on
533: \citet{spitzer78}. These reactions are the same as our previous paper\citep{MS04}.
534: Here we stress the importance to include above rare elements such as Li,
535: because the coupling of magnetic field with gas can be attained by
536: very low fractional abundance of electrons and ions \citep{MS04}.
537:
538: \subsection{Drift velocity of magnetic field}
539: Magnetic field is dissipated from star-forming gas via ohmic loss
540: and ambipolar diffusion.
541: We assess the drift velocity $v_{{\rm B}x}$ of the field lines due to
542: these two processes, which is compared to the accretion velocity of gas.
543: We evaluate the drift velocity following the formulation by
544: \citet{nakano86}.
545: There are two important quantities which characterize
546: these diffusion processes. They are $\tau_\nu$ and $\omega_\nu$ which denote
547: the viscous damping time of the relative velocity of charged particle $\nu$
548: to the neutral particles, and the cyclotron frequency of
549: the charged particle $\nu$, respectively.
550: Then, $\tau_\nu$ is expressed as
551: \begin{equation}
552: \tau_\nu = \frac{\rho_\nu}{\mu_{\rm \nu n}
553: n_\nu n_{\rm n}\langle\sigma v\rangle_{\rm \nu n}},
554: \end{equation}
555: where $\mu_{\rm \nu n}$ is the reduced mass, $n_\nu$, $n_{\rm n}$,
556: and $\rho_\nu$ are, the mean number density for
557: the charged particle $\nu$, the neutral particle ${\rm n}$, and the mass
558: density of charged particle $\nu$, respectively. The averaged momentum-transfer rate coefficient for a
559: particle $\nu$ colliding with a neutral particle is expressed by $\langle \sigma v \rangle_{\rm \nu n}$ .
560: We use the empirical formulae for the momentum-transfer rate coefficients
561: \citep{kamaya00,sano00}.
562:
563: According to \citet{nakano86}, the drift velocity
564: is given by
565: \begin{equation}
566: v_{{\rm B}x} = \frac{A_1}{A}\frac{1}{c}(\Vec{j}\times\Vec{B})_x,
567: \label{eq:diffusion_velocity}
568: \end{equation}
569: where
570: \begin{eqnarray}
571: A = A_1^2 + A_2^2, \label{eq:A}\\
572: A_1 = \sum_\nu \frac{\rho_\nu \tau_\nu \omega_\nu^2}
573: {1 + \tau_\nu^2 \omega_\nu^2}, \label{eq:A1}\\
574: A_2 = \sum_\nu \frac{\rho_\nu \omega_\nu}
575: {1 + \tau_\nu^2 \omega_\nu^2}, \label{eq:A2}
576: \end{eqnarray}
577: $\Vec{B}$ is the mean magnetic field in the primordial cloud,
578: the suffix $x$ means $x$ direction component in local Cartesian
579: coordinates where the z direction is taken as the direction of
580: $\Vec{B}$.
581: We replace
582: $(1/c)(\Vec{j} \times \Vec{B})_x$ in equation
583: (\ref{eq:diffusion_velocity}) with the mean magnetic force
584: $B^2/4 \pi r$, where $B$ is the mean field strength in the cloud,
585: $r$ is the radius of the cloud at which we assess the drift velocity.
586:
587: \subsection{Initial Conditions}
588: We consider primordial star-forming gas clouds that formed in the
589: mini-halos with $\sim 10^6\;\MO$.
590: We set uniform and spherical gas cloud with $n_{\rm H}=10^3\;{\rm
591: cm^{-3}}$, $T_0=250\;{\rm K}$, and $M=10^4\;\MO$.
592: Such clouds are commonly found 3D cosmological simulations
593: \citep{bromm99,abel00,abel02,bromm02,Yoshida03,oshea06}.
594: We also use the cosmological abundance given by \citet{galli98},
595: as the initial fraction of the chemical compositions.
596: The initial magnetic field strength is expected to be very weak,
597: however, its magnitude is still under discussion. Thus, we regard the
598: field strength as a free parameter of the calculations.
599:
600: \subsection{Switching from Run-away Phase to Accretion Phase}
601: The run-away phase and the subsequent accretion phase have to be treated
602: in different formulation from each other as described in subsections
603: \ref{runaway} and \ref{accretionphase}. Thus, we have to switch the
604: scheme from the method in run-away phase to that in
605: accretion phase. In the light of physical arguments,
606: two phases should be switched when the central run-away
607: collapse is stopped due to thermal pressure or centrifugal force.
608: In our calculation, we switch the scheme when the
609: sonic point of the accreting flow
610: go inside a certain small radius. We set this radius to be
611: $r_{\rm sw}=700\;\RO$,
612: which is comparable to the disk size when the accreted
613: mass is comparable to $\sim 1\MO$\citep{TM04}. It is also worth noting
614: that the central density of the core at switching is $\sim 10^{16}~{\rm cm^{-3}}$.
615: %Because of the supersonic accretion in the envelope, above switching
616: %guarantees that the artificial effects due to the introduction of
617: %$r_{\rm sw}$ never affect the thermal and dynamical evolution of the
618: %accreting matter.
619:
620:
621:
622: \begin{figure}[htb]
623: \begin{center}
624: \FigureFile(12cm,10cm){fig2.eps}
625: %\includegraphics[width=0.7\linewidth]{fig1.eps}
626: \caption{
627: The distribution of the main species, $\rm e^-$,
628: $\rm H^+$, H, $\rm H_2$, $\rm H_2^+$, $\rm H_3^+$,
629: Li, and $\rm Li^+$ are plotted.
630: The vertical axis denotes the fractional abundance $y_i$
631: of the above species, and the horizontal axis
632: denotes the radius from the core center.
633: Left panel represents the snapshot when the central density satisfies $n_{\rm H,c}=10^{12}\;\rm cm^{-3}$,
634: whereas right panel shows the results at $n_{\rm H,c} = 10^{16}\;\rm cm^{-3}$.
635: }
636: \label{fig:chemi_runaway}
637: \end{center}
638: \end{figure}
639: \begin{figure}[bht]
640: \begin{center}
641: \FigureFile(12cm,7cm){fig3.eps}
642: %\includegraphics[width=0.7\linewidth]{fig2.eps}
643: \caption{Drift velocity $v_{{\rm B}x}$ as a function
644: of the radius $r$ and the field strength $B$ at
645: the central density .
646: Contour map of $v_{{\rm B}x}/u$ is shown on $r-B$ plane for $n_{\rm H,c} = 10^{16}\;{\rm cm^{-3}}$
647: Solid curves represent the constant loci along which $\log(v_{{\rm
648: B}x}/u)$ equal to the values labeled on the curve.
649: The dashed curve represents the critical field strength $B_{\rm cr}$
650: given by the equation (\ref{eq:critical field}).}
651: \label{fig:rdiffu_runaway}
652: \end{center}
653: \end{figure}
654:
655: \section{RESULTS}
656: \label{results}
657: \subsection{Runaway Collapse Phase}
658: \subsubsection{The distribution of the chemical abundances}
659: Two snapshots of $y_i$ distribution during the run-away
660: collapse phase are shown in Figure \ref{fig:chemi_runaway}.
661: The left panel shows the distribution
662: at the time when the central density $n_{\rm H,c}$
663: reaches $10^{12}\;{\rm cm^{-3}}$,
664: whereas the right panel represents the distribution
665: at $n_{\rm H,c} = 10^{16}\;{\rm cm^{-3}}$.
666: The horizontal axis denotes the radius from the center
667: of the primordial gas cloud $r\;[\rm cm]$, and
668: the vertical axis is the fractional abundances $y_i$
669: of each species.
670: Since the central density increases as the collapse proceeds,
671: right panel represents the later epoch than that of the left.
672:
673:
674: Roughly speaking, in both of the panels, fraction of electrons ($y_{\rm
675: e}$) in the inner
676: region are smaller than that in the outer part. These results could be
677: simply understood since the recombination process in the inner dense
678: region proceeds faster than in the outer envelope. We also find the opposite
679: behavior in the inner most region of right panel,
680: at which $y_{\rm e}$ increases as $r$ decreases, because of the
681: collisional ionization. It is also worth noting that $\rm Li$ becomes the
682: main provider of electrons around $r\sim 10^{14}{\rm cm}$ at $n_{\rm H,c} = 10^{16}\;{\rm cm^{-3}}$. Because of the combined effects of these two
683: (collisional ionization \& presence of Li), $y_{\rm e}$ never gets lower
684: than $10^{-11}$ as far as we consider $n_{\rm H,c} < 10^{16}\;{\rm
685: cm^{-3}}$ in run-away phase.
686: \subsubsection{Drift velocities in run-away collapse phase}
687: \label{drift_runaway}
688: In Figure \ref{fig:rdiffu_runaway},
689: we show the ratio of the drift velocity $v_{{\rm B}x}$ to the
690: infall velocity $u$
691: when $n_{\rm H,c} =10^{16}\;{\rm cm^{-3}}$ is satisfied.
692:
693: The vertical axis shows the magnetic field strength, whereas the
694: horizontal axis shows the distance from the cloud center.
695: The solid curves in Figure \ref{fig:rdiffu_runaway}
696: show the contours of $\log_{10}(v_{{\rm B}x}/u)$.
697: The magnetic fields are dissipated in
698: the region where $\log_{10}(v_{{\rm B}x}/u)>0$,
699: in contrast, the fields are frozen to the gas
700: in the region where $\log_{10}(v_{{\rm B}x}/u)<0$.
701: The dashed curve in Figure \ref{fig:rdiffu_runaway}
702: is the critical field strength $B_{\rm cr}$
703: that is defined by the equation
704: \begin{equation}
705: \frac{B_{\rm cr}^2}{4\pi r} = \frac{GM(r)\rho(r)}{r^2}.
706: \label{eq:critical field}
707: \end{equation}
708: Note that since we are interested in the collapsing gas cloud,
709: the magnetic force needs to be weaker than the gravitational force.
710: Hence, our calculations are valid in the region
711: where the field strength in the cloud satisfies $B<B_{\rm cr}$.
712: We find clearly from Figure \ref{fig:rdiffu_runaway}
713: that the frozen-in condition $v_{{\rm B}x}/u<1$ is almost
714: always satisfied as long as $B$ is less than the critical
715: field strength $B_{\rm cr}$. We also find the basically same results for
716: other snapshots, i.e. the drift velocity is smaller than the infall
717: velocity anytime and anywhere if $B<B_{\rm cr}$.
718:
719: In addition, if $B<B_{\rm cr}$ is satisfied at some initial time and
720: position $(t_0,r_0)$, this
721: condition also holds at some later epoch
722: $(t,r(t))$. This statement is proved as follows:
723: Combining the magnetic flux conservation equation
724: \begin{equation}
725: 2\pi r_0 dr_0 B(t_0,r_0) = 2\pi r(t) dr B(t,r(t))
726: \end{equation}
727: and the mass conservation law given in equation (\ref{eq:continuity at
728: accretion phase}), we obtain
729: \begin{equation}
730: B(t,r(t))=B(t_0,r_0) \frac{\rho(t,r(t)) r(t)}{\rho(t_0,r_0)r_0},
731: \label{eq:B frozen}
732: \end{equation}
733: Using equations (\ref{eq:critical field}) and (\ref{eq:B frozen}) we
734: have
735: \begin{equation}
736: \frac{B(t,r(t))}{B_{\rm cr}(t,r(t))}=\frac{B(t_0,r_0)}{B_{\rm
737: cr}(t_0,r_0)}\left(\frac{\rho(t,r(t)) r^3\left(t\right)}{\rho\left(t_0,r_0\right)r_0^3}\right)^{1/2}
738: \label{eq:B ratio}
739: \end{equation}
740: The second term in right side is always less than or equals to unity
741: in the run-away collapsing cloud,
742: since $\rho\propto r^{-2}$ in the envelope, and
743: $\rho\propto r^{-3}$ in the core.
744: Thus, if ${B(t_0,r_0)}/{B_{\rm cr}(t_0,r_0)} < 1$ is satisfied,
745: $B$ is less than $B_{\rm cr}$ throughout the collapse.
746: Remark that we assume the flux conservation as equation (\ref{eq:B
747: frozen}), which gives the maximal field strength. Therefore, above
748: inequality ($B<B_{\rm cr}$) also holds even if the magnetic flux is dissipated.
749:
750: In summary, it is concluded that the magnetic fields
751: are always frozen to the whole cloud
752: in the course of run-away collapse phase, if the magnetic force is much
753: smaller than gravitational force at the beginning of the collapse.
754:
755: We also remark that the slight increase due to collisional
756: ionization at $n_{\rm H,c} =10^{16}\;{\rm cm^{-3}}$ implies that $y_{\rm
757: e}$ will become larger as the central density increases up to $n_{\rm
758: H,c} > 10^{16}\;{\rm cm^{-3}}$, since the temperature gets higher as the
759: density increases for $\gamma=1.09$ polytropic gas. As a result, the
760: magnetic field is expected to be frozen to the gas even for $n_{\rm
761: H,c} > 10^{16}\;{\rm cm^{-3}}$ in accretion phase.
762: \begin{figure}[tbh]
763: \begin{center}
764: \FigureFile(13cm,7cm){fig4.eps}
765: %\includegraphics[width=0.7\linewidth]{fig3.eps}
766: \caption{The evolutionary sequences of the primordial gas cloud
767: in accretion phase are shown. Four panels (a)-(d) show the spatial
768: distribution of following physical variables as functions of radius:
769: (a) number density $n_{\rm H}$,
770: (b) temperature $T$,
771: (c) velocity $u$, and
772: (d) electron fraction $y_{\rm e}$.
773: Five time sequences (0-4) are plotted. Accreted central mass
774: $M_{\rm c}(t)$ is used as a clock.
775: Corresponding mass at the stages are:~
776: 0:$M_{\rm c}(t) = 8.87\times10^{-2}\;\MO$,
777: 1:$M_{\rm c}(t) = 1.58\;\MO$,
778: 2:$M_{\rm c}(t) = 12.7\;\MO$,
779: 3:$M_{\rm c}(t) = 50.2\;\MO$,
780: 4:$M_{\rm c}(t) = 100\;\MO$.
781: }
782: \label{fig:evol_sequences}
783: \end{center}
784: \end{figure}
785: \begin{figure}[tbh]
786: \begin{center}
787: \FigureFile(12cm,7cm){fig5.eps}
788: %\includegraphics[width=0.7\linewidth]{fig4.eps}
789: \caption{The radial distributions of
790: the fractional abundances of the main species,
791: e, $\rm H^+$, H, $\rm H_2$, $\rm H^-$, $\rm H_3^+$, Li,
792: and $\rm Li^+$ in the accretion phase. Two panels correspond to the
793: snapshots at $M_{\rm c}(t)=12.7\;\MO$(left), and $M_{\rm c}(t)=100\;\MO$(right).
794: }
795: \label{fig:chemi_accretion}
796: \end{center}
797: \end{figure}
798:
799: %\begin{figure}[htb]
800: %\begin{center}
801: %\FigureFile(12cm,7cm){fig6.eps}
802: %\includegraphics[width=0.7\linewidth]{fig5.eps}
803: %\caption{The same as Fig.\ref{fig:rdiffu_runaway},
804: %except that the coutours are plotted in the accretion phase characterized
805: % at $M_{\rm c}$ =$1.58\;\MO$.}
806: %\label{fig:rdiffu_accretion_158}
807: %\end{center}
808: %\end{figure}
809:
810: \begin{figure}[htb]
811: \begin{center}
812: \FigureFile(12cm,7cm){fig6.eps}
813: %\includegraphics[width=0.7\linewidth]{fig6.eps}
814: \caption{The same as Fig.\ref{fig:rdiffu_runaway},
815: except that the contours are plotted in the accretion phase characterized
816: at $M_{\rm c}$ =$12.7\;\MO$.}
817: %\caption{The same as Fig.\ref{fig:rdiffu_accretion_158},
818: %except $M_{\rm c}=12.7\;\MO$.}
819: \label{fig:rdiffu_accretion_127}
820: \end{center}
821: \end{figure}
822:
823: \begin{figure}[htb]
824: \begin{center}
825: \FigureFile(12cm,7cm){fig7.eps}
826: %\includegraphics[width=0.7\linewidth]{fig7.eps}
827: \caption{The same as Fig.\ref{fig:rdiffu_accretion_127},
828: except $M_{\rm c}=100\;\MO$.}
829: \label{fig:rdiffu_accretion_100}
830: \end{center}
831: \end{figure}
832:
833:
834: \subsection{Accretion Phase}
835: \subsubsection{Evolution of gas and chemical species in accretion phase}
836:
837:
838: The evolutionary sequences of number density $n_{\rm H}$,
839: temperature $T$, velocity $u$, and
840: electron fraction $y_{\rm e}$ in the accretion phase
841: are illustrated in Figure \ref{fig:evol_sequences}.
842:
843: Upper left panel illustrates the evolution of density profile. The solid
844: curves labeled as 0-4 corresponds to the epoch at which the central
845: accreted mass equals to $8.87\times10^{-2}\;\MO,1.58\;\MO,12.7\;\MO,50.2\;\MO,
846: 100\;\MO$, respectively.
847: It is clear that the density for $r\to 0$
848: decreases as the collapse proceeds, which is basically the same
849: feature found in similarity solution of polytropic cloud in accretion phase.
850: %The dotted line in the upper left panel in Figure \ref{fig:evol_sequences}
851: %shows the slope of density gradient $\rho \propto r^{-3/2}$ which is
852: %expected in the polytropic similarity solutions for $r\to 0$.
853:
854: Because of the high accretion velocity at later snapshots (lower left),
855: the gas is
856: heated efficiently by adiabatic compression. As a result, temperature
857: gets higher as the collapse proceeds, except at the beginning of the
858: accretion phase, when the gas is cooled efficiently due to the enhanced
859: fraction of H$_2$ molecules.
860:
861: The ionization degree (electron fraction $y_{\rm e}$, lower right) also increases as
862: the collapse proceeds (Figures.\ref{fig:evol_sequences} and
863: \ref{fig:chemi_accretion}), due to the higher temperature for later snapshots.
864: As a result, $y_{\rm e}$ never gets lower than $10^{-11}$. We also find
865: that Li is not so important as was in the final phase of
866: run-away phase, since the hydrogen win back the position of chief
867: provider of electrons in the accretion phase( Figure \ref{fig:chemi_accretion}).
868:
869: \subsubsection{Drift velocities in accretion phase}
870: As briefly discussed in section \ref{intro}, the ambipolar diffusion
871: velocity is proportional to the inverse square of gas density.
872: Its dependence on various physical quantities are described as
873: $v_{{\rm B}x}^{\rm amb}
874: \propto y_{\rm e}^{-1}\rho^{-2}B^2r^{-1}$ (see equations (\ref{eq:diffusion_velocity})-(\ref{eq:A2})).
875: On the other hand, the
876: accretion velocity scales as $u\propto (M_{\rm c}/r)^{1/2}$ for $r\to 0$.
877: Since we consider the accretion phase starting from the final phase of
878: run-away collapse with $\gamma=1.09$, the density profile and the
879: central mass in accretion phase also depend on $\gamma$.
880: According to \citet{suto88}, we have
881: \begin{eqnarray}
882: \rho(t,r)&\propto& t^{(2-3\gamma)/2}r^{-3/2}~~~~~~{\rm for~}r\to 0\\
883: M_{\rm c}(t)&\propto& t^{4-3\gamma}
884: \end{eqnarray}
885: Thus, the limiting behaviour of the infall velocity is described as
886: \begin{eqnarray}
887: u(t,r)\propto t^{(4-3\gamma)/2} r^{-1/2} ~~~~~~{\rm for~}r\to 0
888: \end{eqnarray}
889:
890: Combining above set of equations, we have the dependence of the ratio
891: $v_{{\rm B}x}^{\rm amb}/u$ on $t$:
892: \begin{equation}
893: \frac{v_{{\rm B}x}^{\rm amb}}{u} \propto y_{\rm e}^{-1}B^2 t^{(9\gamma-8)/2}r^{5/2}
894: \end{equation}
895: This equation indicates that the ratio keeps growing in accretion phase
896: for fixed $r,B$ and
897: $y_{\rm e}$, since $9\gamma -8$ is positive for $\gamma =1.09$.
898: In reality, however, $y_{\rm e}$ increases rapidly as the accretion
899: proceeds (see Figures.\ref{fig:evol_sequences} and
900: \ref{fig:chemi_accretion}), and it offsets the increase of $v_{{\rm B}x}^{\rm amb}/u$.
901:
902: Figures
903: %\ref{fig:rdiffu_accretion_158}
904: \ref{fig:rdiffu_accretion_127} and \ref{fig:rdiffu_accretion_100}
905: illustrate
906: the contours of $\log_{10}(v_{{\rm B}x}/u)$ on $r-B$ plane for two
907: epochs.
908: Two figures correspond to the snapshots
909: when the central accreted mass satisfies
910: %$M_{\rm c} = 1.58\;\MO$,
911: $M_{\rm c} = 12.7\;\MO$ and $M_{\rm c} = 100\;\MO$, respectively.
912: The notations are same as Figure \ref{fig:rdiffu_runaway}.
913: It is clear that the drift velocity is always smaller than the accretion
914: velocity in two snapshots, as long as $B<B_{\rm cr}$.
915: In fact, we find that this is true all through the calculations.
916: Besides, the equation (\ref{eq:B ratio}) holds also in the accretion phase.
917: Considering that the density profile in the accretion phase is $\rho\propto
918: r^{-1.5\sim -2}$,
919: $B$ is less than $B_{\rm cr}$ throughout the collapse, if ${B(t_0,r_0)}/{B_{\rm cr}(t_0,r_0)} < 1$, as discussed in \S\ref{drift_runaway}.
920: In other words, magnetic force is always negligible if it can be ignored
921: in the beginning of the collapse.
922:
923: Thus, it is concluded that the dissipation of magnetic flux
924: is negligible throughout the mass accretion phase, as well as the
925: run-away collapse phase.
926: %%%%%
927: \section{DISCUSSION}
928: \label{discussion}
929: %%%%%
930: %\subsection{Jets and MRI in population III star formation}
931: We confirmed that the magnetic field is frozen to the star-forming primordial
932: gas cloud even in the accretion phase. Here we discuss the possibility
933: of jet formation and activation of MRI considering the magnetic field
934: strength brought into the accretion disk surrounding the protostar.
935:
936: The magnetic field strength brought into the accretion disk is assessed
937: under the frozen-in condition as follows:
938:
939: \begin{equation}
940: B_{\rm disk} = B_0\left(\frac{R}{r_{\rm disk}}\right)^2.
941: \end{equation}
942: Here $B_0$ is the initial field strength, $r_{\rm disk}$ denotes the disk radius, whereas $R$ describes the
943: initial radius within which includes the
944: total mass of the disk-star star system
945: $M_{\rm *disk}$:
946: \begin{equation}
947: R = (3M_{\rm *disk}/4\pi \rho_0)^{1/3},
948: \end{equation}
949: where $\rho_0$ represents the initial density of the cloud.
950: $r_{\rm disk}$ can be evaluated by equation (15) in \citet{TM04}:
951: \begin{equation}
952: r_{\rm disk} \simeq 66.4 {\rm AU}\left(\frac{f_{\rm
953: Kep}}{0.5}\right)^2\left(\frac{M_{\rm *disk}}{10 \MO}\right)^{9/7}
954: \end{equation}
955: where $f_{\rm Kep}$ denotes the ratio of the rotation velocity to the
956: Kepler velocity of accreting matter, which is found to be $\sim 0.5$ in
957: \citet{abel02}. Thus, we have the magnetic field strength in the disk as
958: follows:
959:
960: \begin{equation}
961: B_{\rm disk} \simeq 7.5\times 10^{-10} {\rm G}
962: \left(\frac{B_0}{3.7\times 10^{-16}{\rm G}}\right)
963: \left(\frac{n_{\rm H}}{10^3{\rm cm^{-3}}}\right)^{-2/3}
964: \left(\frac{f_{\rm Kep}}{0.5}\right)^{-4}
965: \left(\frac{M_{\rm *disk}}{10 \MO}\right)^{-40/21}
966: \label{eq:Bdisk}
967: \end{equation}
968:
969: Several possibilities to generate cosmological seed magnetic field have
970: been proposed so far. Most of the mechanisms predict $B_{\rm IGM}\lesssim
971: 10^{-19}$G\citep{widrow02}, except the magnetic field generated by
972: radiation transfer effects of powerful ionizing sources such as quasars
973: or first stars \citep{langer03}. They suggests the possibility to
974: generate coherent magnetic field with $B_{\rm IGM} \sim 10^{-11}$G.
975: Since the magnetic field is frozen to the primordial gas at low
976: densities ($n_{\rm H} < 10^3{\rm cm^{-3}}$), $B_0$ at $n_{\rm H} =
977: 10^3{\rm cm^{-3}}$ can be evaluated as
978: \begin{equation}
979: B_0 = 3.7\times 10^{-16} {\rm G}
980: \left(\frac{B_{\rm IGM}}{10^{-19}{\rm G}}\right)
981: \left(\frac{n_{\rm H}}{10^3{\rm cm^{-3}}}\right)^{2/3}
982: \left(\frac{1+z}{20}\right)^2,\label{eq:B0}
983: \end{equation}
984:
985: Recently, 3-dimensional MHD simulations on primordial star formation
986: have been performed by \citet{machida06}, assuming ideal MHD condition is
987: always satisfied in the collapsing gas. In fact, our present results
988: guarantee this hypothesis.
989: They found that
990: the protostellar jet is driven in primordial environment
991: if $B_0\gtrsim 10^{-9}$G at $n_{\rm H} = 10^3{\rm cm^{-3}}$.
992: Comparing this condition with equation (\ref{eq:B0}), it is concluded
993: that 1) jets could be driven in first star forming clouds if the seed
994: field is generated by the mechanism proposed by \citet{langer03}, 2)
995: whereas the other mechanisms cannot generate the seed field enough to
996: drive the jets.
997:
998: MRI can be activated in the accretion disk in case the magnetic field in
999: the disk is larger than a critical value \citep{TB04}:
1000: \begin{equation}
1001: B_{\rm disk} \gtrsim 1.1\times 10^{-4}{\rm G} \left(\frac{M_{\rm *disk}}{10\ M_\odot}\right)^{1/4}
1002: \left(\frac{T}{10^4\ {\rm K}}\right)^{-3/4}
1003: \left(\frac{\ln\Lambda}{10}\right)^{1/2}
1004: \left(\frac{\rho_{\rm disk}}{5\times 10^{-10}\ {\rm g\;cm^{-3}}}\right)^{1/2}
1005: \left(\frac{r}{600\ R_\odot}\right)^{-3/4},
1006: \label{eq:BMRI}
1007: \end{equation}
1008: This threshold is assessed by the confrontation between the growth rate
1009: of MRI and the ohmic dissipation rate. Combining equations
1010: (\ref{eq:Bdisk})- (\ref{eq:BMRI}),
1011: we find that MRI is activated in case the seed field satisfy
1012: \begin{eqnarray}
1013: B_{\rm IGM} \gtrsim 1.5 \times 10^{-14}{\rm G}&~&
1014: \left(\frac{1+z}{20}\right)^{-2}
1015: \left(\frac{f_{\rm Kep}}{0.5}\right)^4
1016: \left(\frac{M_{\rm *disk}}{10 \MO}\right)^{2.155} \nonumber \\
1017: &\times&\left(\frac{T}{10^4\ {\rm K}}\right)^{-3/4}
1018: \left(\frac{\ln\Lambda}{10}\right)^{1/2}
1019: \left(\frac{\rho_{\rm disk}}{5\times 10^{-10}\ {\rm g\;cm^{-3}}}\right)^{1/2}
1020: \left(\frac{r}{600\ R_\odot}\right)^{-3/4}
1021: .
1022: \end{eqnarray}
1023: Therefore, MRI is driven only if the seed field generation mechanism by
1024: the transfer effects of ionizing radiation works.
1025: Based upon these arguments, we
1026: emphasize that the mechanism proposed by
1027: \citet{langer03} should be scrutinized since their results still based
1028: upon the argument of order-estimation.
1029:
1030: %SILK
1031:
1032: \section{SUMMARY}
1033: \label{summary}
1034: In this paper, we investigate
1035: the dissipation of magnetic flux in star-forming primordial gas cloud.
1036: We solve non-equilibrium chemical reaction equations, coupled with
1037: thermal and dynamical evolution of the collapsing cloud all through the
1038: run-away phase as well as the mass accretion phase. Thus, we obtain the
1039: detailed evolution of ionized fraction of the gas, which enables us to
1040: assess the coupling between gas and magnetic field.
1041:
1042: As a result, we find that the magnetic field is
1043: basically
1044: frozen to the gas anywhere in {\it collapsing } star-forming primordial clouds at any time.
1045: Based upon this result, we find the cosmological seed magnetic
1046: field generated by most of the mechanisms proposed so far is not
1047: sufficient to form jets as well as to activate MRI in the star-forming
1048: cloud. Only one mechanism proposed by \citet{langer03} is able to create
1049: sufficient field strength.
1050:
1051: \bigskip
1052:
1053: We thank Kazu Omukai for stimulating discussions. Noriaki Shibazaki and
1054: Ken Ohsuga are acknowledged for continuous encouragement.
1055: The analysis has been
1056: made with computational facilities at Rikkyo University.
1057: This work was supported in part by Ministry of Education, Culture,
1058: Sports, Science, and Technology (MEXT), Grants-in-Aid, Specially
1059: Promoted Research 16002003 and Young Scientists (B) 17740110.
1060:
1061: \appendix
1062: \section{Density distribution of free-falling matter}
1063: \label{appendix_a}
1064: In this Appendix, we derive the explicit formula
1065: for $\displaystyle\left(\frac{\partial r}{\partial r_0}\right)_t$ which
1066: gives the density distribution of free-falling matter in equation (\ref{eq:density in accretion phase}).
1067:
1068: Equation of motion (\ref{eq:eq. of motion in accretion phase}) is integrated as
1069: \begin{equation}
1070: \frac{1}{2}\dot{r}^2=\frac{GM_0}{r} - \mathcal{E}_0,
1071: \quad \mathcal{E}_0=\frac{GM_0}{r_0} - \frac{1}{2}u_0^2,
1072: \quad (\mathcal{E}_0>0),
1073: \label{eq:energy integral}
1074: \end{equation}
1075: where $u_0$ is the velocity of a fluid element at $(t_0,r_0)$ where $r_0$
1076: denotes the position of the element at some initial time $t_0$.
1077: A solution of this equation is given as
1078: \begin{eqnarray}
1079: r(\alpha,\alpha_0) &=& \frac{GM_0}{2\mathcal{E}_0}(1-\cos\alpha), \\
1080: t(\alpha,\alpha_0) &=& t_0 - \frac{GM_0}{(2\mathcal{E}_0)^{3/2}}
1081: \left[(\alpha - \sin\alpha) - (\alpha_0 - \sin\alpha_0)\right],
1082: \end{eqnarray}
1083: where $\alpha$ is the so-called development angle, and $\alpha_0$
1084: is its value at $t=t_0$. $\alpha_0$ also satisfies following relation
1085: \begin{equation}
1086: r_0 = \frac{GM_0}{2\mathcal{E}_0}(1-\cos\alpha_0).
1087: \end{equation}
1088:
1089: Using above relations, $(\partial r/\partial r_0)_t$ is derived as
1090: \begin{equation}
1091: \left(\frac{\partial r}{\partial r_0}\right)_t
1092: = \left(\frac{\partial r}{\partial M_0}\right)_{\mathcal{E}_0,\alpha}
1093: \left(\frac{\partial M_0}{\partial r_0}\right)_t
1094: + \left(\frac{\partial r}{\partial \mathcal{E}_0}\right)_{\alpha,M_0}
1095: \left(\frac{\partial \mathcal{E}_0}{\partial r_0}\right)_t
1096: + \left(\frac{\partial r}{\partial \alpha}\right)_{M_0,\mathcal{E}_0}
1097: \left(\frac{\partial \alpha}{\partial r_0}\right)_t,
1098: \end{equation}
1099: where
1100: \begin{eqnarray}
1101: \left(\frac{\partial r}{\partial M_0}\right)_{\mathcal{E}_0,\alpha}
1102: &=& \frac{G}{2\mathcal{E}_0}(1-\cos\alpha) = \frac{r}{M_0}, \\
1103: \left(\frac{\partial r}{\partial \mathcal{E}_0}\right)_{\alpha,M_0}
1104: &=& -\frac{GM_0}{2\mathcal{E}_0^2}(1-\cos\alpha)=-\frac{r}{\mathcal{E}_0}, \\
1105: \left(\frac{\partial r}{\partial \alpha}\right)_{M_0,\mathcal{E}_0}
1106: &=& \frac{GM_0}{2\mathcal{E}_0}\sin\alpha, \\
1107: \frac{d\alpha_0}{dr_0}
1108: &=& \left(1 - \frac{r_0}{M_0}\frac{dM_0}{dr_0}
1109: + \frac{r_0}{\mathcal{E}_0}\frac{d\mathcal{E}_0}{dr_0}\right)
1110: /\left(\frac{GM_0}{2\mathcal{E}_0}\sin\alpha_0\right), \\
1111: \left(\frac{\partial M_0}{\partial r_0}\right)_t
1112: &=& \frac{dM_0}{dr_0} = 4\pi r_0^2\rho_0, \\
1113: \left(\frac{\partial \mathcal{E}_0}{\partial r_0}\right)_t
1114: &=& \frac{d\mathcal{E}_0}{dr_0}
1115: = \frac{G}{r_0}\frac{dM_0}{dr_0}-\frac{GM_0}{r_0^2}-u_0\frac{du_0}{dr_0}, \\
1116: \left(\frac{\partial \alpha}{\partial r_0}\right)_t
1117: &=& \frac{r_0}{r}\frac{d\alpha_0}{dr_0}
1118: +\frac{\left(t-t_0\right)\sqrt{2\mathcal{E}_0}}{rM_0}\frac{dM_0}{dr_0}
1119: -\frac{3\left(t-t_0\right)}{r\sqrt{2\mathcal{E}_0}}\frac{d\mathcal{E}_0}{dr_0}.
1120: \end{eqnarray}
1121: %\clearpage
1122:
1123:
1124: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1125: %%%%%% (5) References %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1126: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1127: \begin{thebibliography}{}
1128: \bibitem[Abel, Bryan \& Norman(2000)]{abel00}
1129: Abel, T., Bryan, G. L., \& Norman, M. L. 2000, \apj, 540, 39
1130: \bibitem[Abel, Bryan \& Norman(2002)]{abel02}
1131: Abel, T., Bryan, G. L., \& Norman, M. L. 2002, Science, 295, 93
1132: %\bibitem[Becker et al. (2001)]{Becker01}
1133: % Becker, R. H. et al. 2001, \aj, 122, 2850
1134: \bibitem[Bodenheimer et al.(2000)]{bodenheimer00}
1135: Bodenheimer, P., Burkert, A., Klein, R., \& Boss, A. 2000,
1136: Protostars and Planets IV
1137: 2000, eds. V. Mannings, A. P. Boss, and S. S. Russel,
1138: U. of Arizona Press, 675
1139: \bibitem[Bromm, Coppi \& Larson(1999)]{bromm99}
1140: Bromm, V., Coppi, P. S., \& Larson, R. B. 1999, \apj, 527, L5
1141: \bibitem[Bromm, Coppi \& Larson(2002)]{bromm02}
1142: Bromm, V., Coppi, P. S., \& Larson, R. B. 2002, \apj, 564, 23
1143: %\bibitem[Carlberg(1981)]{Carl}
1144: % Carlberg, R. G. 1981, \mnras, 197, 1021
1145: \bibitem[Christlieb et al.(2004)]{Christlieb04}
1146: Christlieb, N. et al. 2004, \apj, 603, 708
1147: \bibitem[Colella \& Woodward(1984)]{CW84}
1148: Colella, P., \& Woodward, P.R. 1984, Jour. Comp. Phys., 54, 174
1149: %\bibitem[Fan et al.(2002)]{Fan02}
1150: % Fan, X. et al. 2002, \aj, 123, 1247
1151: \bibitem[Flower(2002)]{flower02}
1152: Flower, D. R. 2002, \mnras, 333, 763
1153: \bibitem[Frebel et al.(2005)]{Frebel05}
1154: Frebel, A. et al. 2005, \nat, 434, 871
1155: \bibitem[Galli \& Palla(1998)]{galli98}
1156: Galli, D., \& Palla, F. 1998, \aap, 335, 403
1157: \bibitem[Hawley \& Balbus(1992)]{hawley92}
1158: Hawley, J. F., \& Balbus, S. A. 1992, \apj, 400, 595
1159: %\bibitem[Hayashi(1981)]{hayashi81}
1160: % Hayashi, M. 1981, Inst. Plasma Phys. Japan Int. Rep.
1161: % IPPJ-AM-19
1162: \bibitem[Hollenbach \& McKee(1979)]{HM79}
1163: Hollenbachm D., \& McKee,C.F. 1979,\apjs, 41, 555
1164: %\bibitem[Hutchins(1976)]{Hut}
1165: % Hutchins, J. B. 1976, \apj, 205, 103
1166: \bibitem[Iwamoto et al.(2005)]{Iwamoto05}
1167: Iwamoto, N. et al. 2005, Science, 309, 451
1168: \bibitem[Johnson \& Bromm (2006)]{JB06}
1169: Johnson, J.L. \& Bromm, V. 2006, \mnras, 366, 247
1170: \bibitem[Kamaya \& Nishi(2000)]{kamaya00}
1171: Kamaya, H., \& Nishi, R. 2000, \apj, 543, 257
1172: %\bibitem[Kogut et al.(2003)]{Kogut03}
1173: % Kogut, A. et al. 2003, \apjs, 148, 161
1174: %\bibitem[Kulsrud et al.(1997)]{kulsrud97}
1175: % Kulsrud, R. M., Cen, R., Ostriker, J. P., \& Ryu, D.
1176: % 1997, \apj, 480, 481
1177: \bibitem[Langer, Puget \& Aghanim(2003)]{langer03}
1178: Langer, M., Puget, J., \& Aghanim, N. 2003, \prd, 67, 43505
1179: \bibitem[Larson(1969)]{Larson69}
1180: Larson, R. 1969, \mnras, 145, 271
1181: \bibitem[Lenzuni, Chernoff \& Salpeter(1991)]{lenzuni91}
1182: Lenzuni, P., Chernoff, D. F., \& Salpeter, E. E. 1991, \apjs, 76, 759
1183: \bibitem[Lepp, Stancil \& Dalgarno(2002)]{lepp02}
1184: Lepp, S., Stancil, P. C., \& Dalgarno, A. 2002, J. Phys. B, 35, R57
1185: \bibitem[Levermore \& Pomraning(1981)]{LP81}
1186: Levermore,C.D \& Pomraning,G.C. 1981, \apj, 248, L321
1187: %\bibitem[Ma(2002)]{ma02}
1188: % Ma, F. 2002, \mnras, 335, L99
1189: \bibitem[Machida et al.(2006)]{machida06}
1190: Machida, M. N., Omukai, K., Matsumoto, T., \& Inutsuka, S. 2006,
1191: ApJL, 647, 1
1192: \bibitem[Maki \& Susa(2004)]{MS04}
1193: Maki, H., \& Susa, H. 2004, \apj, 609, 467
1194: %\bibitem[Matsuda, Sato \& Takeda(1969)]{MST69}
1195: % Matsuda, T., Sato H., \& Takeda, H. 1969, Prog. Theor. Phys., 42, 219
1196: \bibitem[Nakamura \& Umemura(1999)]{nakamura99}
1197: Nakamura, F., \& Umemura, M. 1999, \apj, 515, 239
1198: \bibitem[Nakamura \& Umemura(2001)]{nakamura01}
1199: Nakamura, F., \& Umemura, M. 2001, \apj, 548, 19
1200: \bibitem[Nakano \& Umebayashi(1986)]{nakano86}
1201: Nakano, T., \& Umebayashi, T. 1986, \mnras, 218, 663
1202: \bibitem[Omukai(2000)]{omukai00}
1203: Omukai, K. 2000, \apj, 534, 809
1204: \bibitem[Omukai \& Nishi (1998)]{omukai98}
1205: Omukai, K., \& Nishi, R. 1998, \apj, 508, 141
1206: \bibitem[O'Shea, Abel, Whalen \& Norman(2005)]{oshea05}
1207: O'Shea, B.~W., Abel, T., Whalen, D., Norman, M.~L. \ 2005, \apj, 628, 5L
1208: \bibitem[O'Shea \& Norman(2006)]{oshea06}
1209: O'Shea, B.~W. \& Norman, M.~L. \ 2006, \apj, 648, 31
1210: \bibitem[O'Shea \& Norman(2007)]{oshea07}
1211: O'Shea, B.~W. \& Norman, M.~L. \ 2007, \apj, 654, 66
1212: %\bibitem[Osterbrock(1961)]{osterbrock61}
1213: % Osterbrock, D. E. 1961, \apj, 134, 270
1214: %\bibitem[Palla, Salpeter \& Stahler(1983)]{PSS83}
1215: % Palla, F., Salpeter, E. E., \& Stahler, S. W. 1983, \apj, 271, 632
1216: \bibitem[Penston(1969)]{Penston69}
1217: Penston, M. V. 1969, MNRAS, 144, 425
1218: %\bibitem[Pudritz \& Silk(1989)]{pudritz89}
1219: % Pudritz, R. E., \& Silk, J. 1989, \apj, 342, 650
1220: %\bibitem[Puy \& Signore(1996)]{Puy}
1221: % Puy, D., \& Signore, M. 1996, \aap, 305, 371
1222: \bibitem[Saigo, Matsumoto \& Umemura(2005)]{saigo05}
1223: Saigo, K, Matsumoto, T \& Umemura, M. 2005, \apj, 615, 65
1224: \bibitem[Sano, Inutsuka \& Miyama(1998)]{sano98}
1225: Sano, T., Inutsuka, S., \& Miyama, S. M. 1998, \apj, 506, L57
1226: \bibitem[Sano et al.(2000)]{sano00}
1227: Sano, T., Miyama, S. M., Umebayashi, T., \& Nakano, T.
1228: 2000, \apj, 543, 486
1229: \bibitem[Sano \& Inutsuka(2001)]{sano01}
1230: Sano, T., \& Inutsuka, S. 2001, \apj, 561, L179
1231: %\bibitem[Shapiro \& Kang(1987)]{shapiro87}
1232: % Shapiro, P. R., \& Kang, H. 1987, \apj, 318, 32
1233: %\bibitem[Songaila(2001)]{songaila01}
1234: % Songaila, A. 2001, \apj, 561, L153
1235: %\bibitem[Sokasian et al.(2004)]{Sokasian04}
1236: % Sokasian, A., Yoshida, N., Abel, T., Hernquist, L., \& Springel, V.
1237: % 2004, \mnras, 350, 47
1238: \bibitem[Spitzer(1978)]{spitzer78}
1239: Spitzer, L. 1978, \textit{Physical Processes in the
1240: Interstellar Medium}, (New York: Wiley Interscience) p.107
1241: \bibitem[Stancil, Lepp \& Dalgarno(1998)]{stancil98}
1242: Stancil, P. C., Lepp, S., \& Dalgarno, A. 1998, \apj, 509, 1
1243: \bibitem[Stone et al.(2000)]{stone00}
1244: Stone, J., Gammie, C., Balbus, S., \& Hawley, J. 2000,
1245: Protostars and Planets IV
1246: 2000, eds. V. Mannings, A. P. Boss, and S. S. Russel,
1247: U. of Arizona Press, 589
1248: %\bibitem[Susa, Uehara \& Nishi (1996)]{SUN96}
1249: % Susa, H., Uehara, H., \& Nishi, R. 1996, Prog. Theor. Phys., 96, 1073
1250: \bibitem[Susa(2007)]{Susa07} Susa, H.\ 2007, ArXiv
1251: Astrophysics e-prints, arXiv:astro-ph/0701172
1252: \bibitem[Suto \& Silk (1988)]{suto88}
1253: Suto, Y., \& Silk, J. 1988, \apj, 326, 527
1254: \bibitem[Tan \& Blackman(2004)]{TB04}
1255: Tan, J. C., \& Blackman, E. G. 2004, \apj, 603, 401
1256: \bibitem[Tan \& Mckee(2004)]{TM04}
1257: Tan, J. C., \& McKee, C. F. 2004, \apj, 603, 383
1258: %\bibitem[Uehara et al.(1996)]{USNY96}
1259: % Uehara, H., Susa, H., Nishi, R., Yamada, M., \& Nakamura, T. 1996,
1260: % \apj, 473, L95
1261: %\bibitem[Vladilo(2002)]{vladilo02}
1262: % Vladilo, G. 2002, \aap, 391, 407
1263: %\bibitem[White et al.(2003)]{white03}
1264: % White, R. L., Becker, R. H., Fan, X., \& Strauss, M. A. 2003,
1265: % \aj, 126, 1
1266: \bibitem[Widrow(2002)]{widrow02}
1267: Widrow, L. M. 2002, Rev. Mod. Phys., 74, 775
1268: \bibitem[Yoshida et al.(2003)]{Yoshida03}
1269: Yoshida, N., Abel, T., Hernquist, L., \& Sugiyama, N. 2003,
1270: \apj, 592, 645
1271: \bibitem[Yoshida et al.(2006)]{Yoshida06}
1272: Yoshida, N., Omukai, K., Hernquist, L. \& Abel, T. 2006, \apj, 652, 6
1273: \end{thebibliography}
1274:
1275:
1276: %\setcounter{figure}{0}
1277:
1278:
1279:
1280:
1281:
1282:
1283:
1284: \end{document}
1285: