1:
2: %% --------------------------------------------------------------------
3: %% Fri Jun 29 07:00:40 2007
4: %% This file was generated automagically from the files
5: %% ms.bbl and ms.tex using
6: %% ./nat2jour.pl
7: %% All citations have been inlined and dependencies on the natbib
8: %% package have been removed so that this file (together with
9: %% ms-aas.bbl) should be suitable for submission to journals with
10: %% the citation styles of ApJ or MNRAS.
11: %% --------------------------------------------------------------------
12:
13: \documentclass[12pt,preprint]{aastex} % for e-submission to ApJ - one column
14:
15: %\documentclass[12pt,preprint2]{aastex} % for e-submission to ApJ - two column
16:
17: %\documentclass{emulateapj} % to make it look like ApJ
18:
19: \usepackage{graphicx}
20:
21: \def\etal{{\sl et al.}}
22: \def\kms{{\rm km/s}}
23: \def\Ms{{\rm M_\odot}}
24: \def\h2{H$_2$}
25:
26: %\citestyle{aa} % correct formatting for ApJ style files
27:
28: \begin{document}
29:
30: \title{Population III star formation in a $\Lambda$CDM universe, \
31: II: Effects of a photodissociating background}
32:
33: \author{Brian W. O'Shea\altaffilmark{1} \& Michael L. Norman\altaffilmark{2}}
34:
35: \altaffiltext{1}{Theoretical Astrophysics (T-6), MS B227, Los Alamos National
36: Laboratory, Los Alamos, NM 87545; bwoshea@lanl.gov}
37:
38: \altaffiltext{2}{Center for Astrophysics and Space Sciences,
39: University of California at San Diego, La Jolla, CA
40: 92093; mlnorman@ucsd.edu}
41:
42: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
43: \begin{abstract}
44: We examine aspects of primordial star formation in the presence of a molecular
45: hydrogen-dissociating ultraviolet background. We compare a set of AMR hydrodynamic
46: cosmological simulations using a single cosmological realization but with a range
47: of ultraviolet background strengths in the Lyman-Werner band. This allows us to
48: study the effects of Lyman-Werner radiation on suppressing \h2 cooling at low
49: densities as well as the high-density evolution of the collapsing cloud core in a
50: self-consistent cosmological framework. We find that the addition of a
51: photodissociating background results in a delay of the collapse of high density
52: gas at the center of the most massive halo in the simulation and, as a result,
53: an increase in the virial mass of this halo at the onset of baryon collapse.
54: We find that, contrary to previous results, Population III star formation is
55: not suppressed for J$_{21} \geq 0.1$, but occurs even with backgrounds as high
56: as J$_{21} = 1$. We find that \h2 cooling leads to collapse despite the
57: depressed core molecular hydrogen fractions due to the elevated \h2 cooling
58: rates at $T=2-5 \times 10^3$ K. We observe a relationship between the strength
59: of the photodissociating background and the rate of accretion onto the evolving
60: protostellar cloud core, with higher LW background fluxes resulting in higher
61: accretion rates. Finally, we find that the collapsing cloud cores in our
62: simulations do not fragment at densities below $n \sim 10^{10}$~cm$^{-3}$
63: regardless of the strength of the LW background, suggesting that Population
64: III stars forming in halos with T$_{vir} \sim 10^4$~K may still form in isolation.
65: \end{abstract}
66:
67: \keywords{cosmology: theory --- galaxies: high-redshift --- stars: formation --- hydrodynamics}
68:
69:
70:
71: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
72: \section{Introduction}\label{intro}
73:
74: The study of the formation of Population III stars in a cosmological context
75: via high-resolution simulations is becoming a mature discipline, with multiple
76: groups finding similar results
77: \markcite{2002ApJ...564...23B, 2004NewA....9..353B, ABN02, oshea07a, 2006ApJ...652....6Y,
78: 2006astro.ph.10174G}({Bromm}, {Coppi}, \& {Larson} 2002; {Bromm} \& {Loeb} 2004; {Abel}, {Bryan}, \& {Norman} 2002; {O'Shea} \& {Norman} 2007; {Yoshida} {et~al.} 2006b; {Gao} {et~al.} 2006). However, all of these calculations operate under a common
79: fundamental assumption -- namely, the absence of an ultraviolet background.
80: Given the abundance of halos in which Population III stars would form and the
81: general consensus that these stars must be massive, one expects
82: that, though only a small fraction of the volume of the universe would be
83: ionized, a significant background of ultraviolet radiation in the Lyman-Werner
84: (LW) band (11.18-13.6 eV), which is capable of photodissociating molecular
85: hydrogen, would be present for the formation of the bulk of Population III stars
86: \markcite{2001ApJ...546..635O,2003ApJ...592..645Y,2001ApJ...548..509M,2005ApJ...629..615W}({Omukai} 2001; {Yoshida} {et~al.} 2003; {Machacek}, {Bryan}, \& {Abel} 2001; {Wise} \& {Abel} 2005).
87: Given that atomic hydrogen is optically thin to this radiation and that it easily
88: destroys molecular hydrogen, this could have a significant impact on the formation
89: of primordial stars. \markcite{2001ApJ...548..509M}{Machacek} {et~al.} (2001) used cosmological AMR simulations to
90: study the formation of primordial stars in the presence of a soft UV background,
91: and later including an x-ray background \markcite{2003MNRAS.338..273M}({Machacek}, {Bryan}, \& {Abel} 2003).
92: They found that the LW background delays the formation
93: of Population III stars and shifts halo formation to higher masses. However, their calculations
94: were not of sufficiently high resolution to give any significant information about
95: the formation of the protostellar cloud at the core of each collapsing cosmological halo.
96: \markcite{2003ApJ...592..645Y}{Yoshida} {et~al.} (2003) performed a suite of large SPH simulations of high redshift
97: structure formation, including two calculations with a soft UV background with
98: $J_{21} = 0.01$ and $0.1$, where $J_{21}$ is the mean intensity of the UV background
99: in the LW band in units of $10^{-21}$~erg s$^{-1}$~cm$^{-2}$~Hz$^{-1}$~sr$^{-1}$. They
100: found that gas cooling is suppressed in the higher LW background case, and predicted that
101: star formation would not occur. On the other hand, \markcite{2002ApJ...569..558O}{Oh} \& {Haiman} (2002) showed
102: analytically that primordial gas in halos with
103: T$_{vir} \ga 10^4$~K can still collapse to high densities in the presence of a strong
104: molecular hydrogen-dissociating background via atomic hydrogen line radiation, and can
105: eventually form H$_2$. They referred to these objects as ``second generation objects" in
106: the sense that although still of primordial composition, their formation pathway is
107: different from the first stars.
108:
109: An issue that needs to be addressed concerns the mode of primordial star formation
110: in halos with T$_{vir} \sim 10^4$~K and above. It is apparent from recent studies using large
111: numbers of simulations that smaller halos, with masses of $\sim 10^5-10^6$~M$_\odot$,
112: appear to always form primordial stars in isolation~\markcite{2006astro.ph.10174G,oshea07a}({Gao} {et~al.} 2006; {O'Shea} \& {Norman} 2007).
113: Is this still the case in halos whose masses are an order of magnitude or more
114: larger? Does a single primordial star form, or several? This is particularly
115: relevant given that many semi-analytic models of structure formation that follow the
116: chemical evolution of structures at high redshift \markcite{2005ApJ...633.1031S,2006MNRAS.369..825S}({Scannapieco} {et~al.} 2005; {Schneider} {et~al.} 2006)
117: suggest that halos of this scale which have not been enriched by metals (and thus will form
118: primordial stars) exist up to at least $z \sim 5$. If this is true, there is expected
119: to be a strong photodissociating background at that time~\markcite{2003ApJ...592..645Y,2005ApJ...629..615W}({Yoshida} {et~al.} 2003; {Wise} \& {Abel} 2005).
120:
121: This paper is the second in a series. In \markcite{oshea07a}{O'Shea} \& {Norman} (2007) (hereafter referred to as Paper I), we
122: examined several aspects of Population III star formation in a $\Lambda$CDM universe, in the
123: absence of an ultraviolet background which photodissociates molecular hydrogen.
124: In this paper, we study Population III star formation in a single cosmological realization
125: varying the strength of the LW
126: background, with the goal of investigating the effect that this background has on the
127: evolution and properties of gas in the halo core (at radii $\ll 1$~pc) and the ultimate
128: fate of the gas at the center of
129: T$_{vir} \sim 10^4$~K halos. We use a single cosmological realization,
130: varying the strength of the photodissociating background over the range
131: suggested by~\markcite{2005ApJ...629..615W}{Wise} \& {Abel} (2005), and look for the cutoff in Population III star formation
132: suggested by~\markcite{2003ApJ...592..645Y}{Yoshida} {et~al.} (2003).
133:
134: In agreement with the findings of \markcite{2001ApJ...548..509M}({Machacek} {et~al.} 2001), we find that the addition of a
135: photodissociating background
136: results in a delay of the collapse of high density gas at the center of the most
137: massive halo in the simulation and, as a result, an increase in the virial mass of this
138: halo at the onset of baryon collapse. We find that, contrary to the results
139: of~\markcite{2003ApJ...592..645Y}{Yoshida} {et~al.} (2003), star formation is not suppressed for J$_{21} \geq 0.1$,
140: but occurs even with backgrounds as high as J$_{21} = 1$.
141: We find that \h2 cooling leads to collapse despite the depressed halo core molecular hydrogen
142: fractions f$_{H2} \sim 10^{-6}$ by two multiplicative effects: (1) the elevated \h2 cooling
143: rates per molecule at $T=2-5 \times 10^3$ K, and (2) time. We find that halo core collapse occurs in the usual
144: way once the gas in the halo core has become dense enough that the cooling time becomes much less than
145: the Hubble time.
146: We also observe a relationship between the strength
147: of the photodissociating background and the rate of accretion onto the evolving
148: protostellar cloud core, with higher LW background fluxes resulting in higher
149: accretion rates. This is a simple consequence of the suppression of molecular
150: hydrogen formation (and thus suppression of cooling) by the photodissociating background, as well as
151: the higher virial temperatures of the more massive
152: halos at the epoch of collapse. This may have implications for the range of Population III stellar
153: masses. Finally, we find that the collapsing halo cores in our simulations do not
154: fragment at densities below $n \sim 10^{10}$~cm$^{-3}$ regardless of the strength of the
155: soft UV background, suggesting that Population III stars forming in halos with
156: T$_{vir} \sim 10^4$~K may still form in isolation.
157:
158: The organization of this paper is as follows. In Section~\ref{methodology} we provide a
159: description of Enzo, the code used to perform the calculations in this paper and of
160: the simulation setup. The results from our simulations are presented in
161: Sections~\ref{results.meanprop} through \ref{results.rephalos}:
162: Section~\ref{results.meanprop} discusses some of the mean halo properties observed in
163: the calculations, Section~\ref{results.evolution} discusses the evolution
164: of the halo core prior to collapse,
165: Section~\ref{results.collapse} compares spherically-averaged halo
166: properties for all simulations at the epoch of collapse,
167: Section~\ref{results.fixred} discusses a variety of halo properties at a fixed
168: redshift, and Section~\ref{results.rephalos} compares the evolution of two representative
169: simulations taken from our ensemble.
170: In Section~\ref{sect.issues} we discuss neglected physics and possible numerical issues,
171: and in Section~\ref{discuss} we discuss some of the results presented in this work and
172: their implications. Finally, we present a summary of the main results in Section~\ref{summary}.
173:
174:
175: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
176: \section{Methodology}\label{methodology}
177:
178: %---------------------------------------------------------------------------
179: \subsection{The Enzo code}\label{enzocode}
180:
181: `Enzo'\footnote{http://lca.ucsd.edu/portal/software/enzo/} is a publicly available, extensively tested
182: adaptive mesh refinement (AMR)
183: cosmology code developed by Greg Bryan and others \markcite{bryan97,bryan99,norman99,oshea04,
184: 2005ApJS..160....1O}({Bryan} \& {Norman} 1997a, 1997b; {Norman} \& {Bryan} 1999; {O'Shea} {et~al.} 2004, 2005b).
185: The specifics of the Enzo code are described in detail in these papers (and references therein),
186: but we present a brief description here for clarity.
187:
188: The Enzo code couples an N-body particle-mesh (PM) solver \markcite{Efstathiou85, Hockney88}({Efstathiou} {et~al.} 1985; {Hockney} \& {Eastwood} 1988)
189: used to follow the evolution of a collisionless dark
190: matter component with an Eulerian AMR method for ideal gas dynamics by \markcite{Berger89}{Berger} \& {Colella} (1989),
191: which allows high dynamic range in gravitational physics and hydrodynamics in an
192: expanding universe. This AMR method (referred to as \textit{structured} AMR) utilizes
193: an adaptive hierarchy of grid patches at varying levels of resolution. Each
194: rectangular grid patch (referred to as a ``grid'') covers some region of space in its
195: \textit{parent grid} which requires higher resolution, and can itself become the
196: parent grid to an even more highly resolved \textit{child grid}. Enzo's implementation
197: of structured AMR places no fundamental restrictions on the number of grids at a
198: given level of refinement, or on the number of levels of refinement. However, owing
199: to limited computational resources it is practical to institute a maximum level of
200: refinement, $\ell_{max}$. Additionally, the Enzo AMR implementation allows arbitrary
201: integer ratios of parent
202: and child grid resolution, though in general for cosmological simulations (including the
203: work described in this paper) a refinement ratio of 2 is used.
204:
205: Since the addition of more highly refined grids is adaptive, the conditions for refinement
206: must be specified. In Enzo, the criteria for refinement can be set by the user to be
207: a combination of any or all of the following: baryon or dark matter overdensity
208: threshold, minimum resolution of the local Jeans length, local density gradients,
209: local pressure gradients, local energy gradients, shocks, and cooling time.
210: A cell reaching
211: any or all of the user-specified criteria will then be flagged for refinement. Once all
212: cells of a given level have been examined, rectangular solid boundaries are determined which
213: minimally
214: encompass the flagged cells on that level. A refined grid
215: patch is then introduced within each such bounding
216: volume, and the results are interpolated to a higher level of resolution.
217:
218: In Enzo, resolution of the equations being solved is adaptive in time as well as in
219: space. The timestep in Enzo is satisfied on a level-by-level basis by finding the
220: largest timestep such that the Courant condition (and an analogous condition for
221: the dark matter particles) is satisfied by every cell on that level. All cells
222: on a given level are advanced using the same timestep. Once a level $L$ has been
223: advanced in time $\Delta t_L$, all grids at level $L+1$ are
224: advanced, using the same criteria for timestep calculations described above, until they
225: reach the same physical time as the grids at level $L$. At this point grids at level
226: $L+1$ exchange baryon flux information with their parent grids, providing a more
227: accurate solution on level $L$. Cells at level $L+1$ are then examined to see
228: if they should be refined or de-refined, and the entire grid hierarchy is rebuilt
229: at that level (including all more highly refined levels). The timestepping and
230: hierarchy rebuilding processes are repeated recursively on every level to the
231: maximum existing grid level in the simulation.
232:
233: Two different hydrodynamic methods are implemented in Enzo: the piecewise parabolic
234: method (PPM) \markcite{Woodward84}({Woodward} \& {Colella} 1984), which was extended to cosmology by
235: \markcite{Bryan95}{Bryan} {et~al.} (1995), and the hydrodynamic method used in the ZEUS magnetohydrodynamics code
236: \markcite{stone92a,stone92b}({Stone} \& {Norman} 1992a, 1992b). We direct the interested reader to the papers describing
237: both of these methods for more information, and note that PPM is the preferred choice
238: of hydro method since it is higher-order-accurate and is based on a technique that
239: does not require artificial viscosity, which smoothes shocks and can smear out
240: features in the hydrodynamic flow.
241:
242: The chemical and cooling properties of primordial (metal-free) gas are followed
243: using the method of \markcite{abel97}{Abel} {et~al.} (1997) and \markcite{anninos97}{Anninos} {et~al.} (1997).
244: This method follows the non-equilibrium evolution of a
245: gas of primordial composition with 9 total species:
246: $H$, $H^+$, $He$, $He^+$, $He^{++}$, $H^-$, $H_2^+$, $H_2$, and $e^-$. The code
247: also calculates
248: radiative heating and cooling following atomic line excitation, recombination,
249: collisional excitation, free-free transitions, molecular line cooling, and Compton
250: scattering of the cosmic microwave background, as well as any of
251: approximately a dozen different models for a metagalactic ultraviolet background that heat
252: the gas via photoionization and/or photodissociation. We model the cooling processes
253: detailed in~\markcite{abel97}{Abel} {et~al.} (1997), but use the~\markcite{1998A&A...335..403G}{Galli} \& {Palla} (1998) molecular
254: hydrogen cooling function. The multispecies rate equations are solved out of
255: equilibrium to properly model situations where, e.g., the cooling time of the gas
256: is much shorter than the hydrogen recombination time.
257: A total of 9 kinetic equations are solved, including 29 kinetic and radiative
258: processes, for the 9 species mentioned above.
259: The chemical reaction equation network is technically challenging to solve due to
260: the huge range of reaction time scales involved; the characteristic creation
261: and destruction time scales of the various species and reactions can differ by
262: many orders of magnitude. As a result, the set of rate equations is extremely
263: stiff, and an explicit scheme for integration of the rate equations can be
264: costly if small enough timestep are taken to keep the network
265: stable. This makes an implicit scheme preferable for such a set of
266: equations, and Enzo solves the rate equations using a method based on a backwards
267: differencing formula (BDF) in order to provide a stable and accurate solution.
268:
269: It is important to note the regime in which this chemistry model is valid. According to
270: \markcite{abel97}{Abel} {et~al.} (1997) and \markcite{anninos97}{Anninos} {et~al.} (1997), the reaction network is valid for temperatures
271: between $10^0 - 10^8$ K. The original model discussed in these two references is only
272: valid up to n$_H \sim 10^4$~cm$^{-3}$. However, addition of the 3-body H$_2$ formation
273: process allows correct modeling of the gas chemistry up until the point where
274: collisionally induced emission from molecular hydrogen becomes an important
275: cooling processes, which occurs at $n_H \sim 10^{14}$~cm$^{-3}$. We do not
276: include heating by molecular hydrogen formation, which will be significant at
277: densities of $\sim 10^8$~cm$^{-3}$ and above, and may affect temperature evolution at
278: these high densities. A further concern is that the optically thin approximation
279: for radiative cooling breaks down beginning at n$_H \simeq 10^{10} - 10^{12}$~cm$^{-3}$.
280: Beyond this point,
281: modifications to the cooling function that take into account the non-negligible
282: opacity of the gas to line radiation from molecular hydrogen must be made, as
283: discussed by \markcite{ripamonti04}{Ripamonti} \& {Abel} (2004). Even with these modifications, a more correct
284: description of the cooling of gas of primordial composition at high densities will
285: require some form of radiation transport, which will greatly
286: increase the cost of the simulations.
287:
288: %---------------------------------------------------------------------------
289: \subsection{Simulation setup}\label{sect.simsetup}
290:
291: The simulations discussed in this paper are set up in a similar way
292: to those in Paper I. A dark matter-only
293: calculation with $128^3$ particles in a three-dimensional simulation volume
294: which is $0.6$~h$^{-1}$~Mpc (comoving) on a side
295: is set up at $z=99$ assuming a ``concordance''
296: cosmological model with no baryons: $\Omega_m = \Omega_{DM} = 0.3$,
297: $\Omega_b = 0.0$, $\Omega_\Lambda = 0.7$, $h=0.7$ (in units of 100 km/s/Mpc),
298: $\sigma_8 = 0.9$, and using an Eisenstein \& Hu power spectrum \markcite{eishu99}({Eisenstein} \& {Hu} 1999)
299: with a spectral index of $n = 1$. The cold dark matter cosmological model
300: is assumed. This calculation is then evolved to
301: $z=15$ using a maximum of four levels of adaptive mesh refinement,
302: refining on a dark matter overdensity of 8.0. At $z=15$, the
303: Hop halo finding algorithm \markcite{eishut98}({Eisenstein} \& {Hut} 1998) is used to find the most massive
304: halo in the simulation.
305:
306: At this point, we generate a new set of initial conditions which contain the same large-scale
307: power as the dark matter-only calculation, but include both dark matter and baryons
308: such that the Lagrangian volume in which the halo in the dark matter-only calculation formed
309: is resolved at high spatial and mass resolution using a series of static
310: nested grids, with a $128^3$
311: root grid and three static nested grids, for an overall effective root grid size of $1024^3$
312: cells.
313: The highest resolution grid is $256^3$ grid cells, and corresponds
314: to a volume $150$~h$^{-1}$ comoving kpc on a side.
315: The dark matter particles in the highest
316: resolution grid are 14.48~h$^{-1}$~M$_\odot$ and the spatial resolution
317: of the highest resolution grid is 586~h$^{-1}$ parsecs (comoving).
318: Previous work shows that this particle mass resolution is adequate
319: to fully resolve the collapse of the halo \markcite{ABN02,oshea07a}({Abel} {et~al.} 2002; {O'Shea} \& {Norman} 2007).
320:
321: All simulations are performed using the adaptive
322: mesh cosmology code Enzo, which is described in Section~\ref{enzocode}.
323: The simulations are started at $z=99$ and allowed to evolve until the collapse
324: of the gas within the center of the most massive halo, assuming the presence of
325: an unevolving soft UV background with intensities in the Lyman-Werner band
326: of J$_{LW} = 0, 10^{-24}, 10^{-23.5}, 10^{-22.5}, 10^{-22}, 10^{-21.67}, 10^{-21.33},$
327: and~$10^{-21}$ ~erg s$^{-1}$~cm$^{-2}$~Hz$^{-1}$~sr$^{-1}$.
328: This range covers a much wider range of parameter space than the results
329: described in \markcite{2001ApJ...548..509M}{Machacek} {et~al.} (2001), and completely encompasses the range of possible J$_{LW}$ values
330: suggested by \markcite{2005ApJ...629..615W}{Wise} \& {Abel} (2005), for a wide range of mean Population III stellar masses.
331: Note that many publications use F$_{LW}$ instead of J$_{LW}$: F$_{LW}$ has units of erg~s$^{-1}$~cm$^{-2}$~Hz$^{-1}$,
332: and thus F$_{LW} = 4 \pi$~J$_{LW}$. We will use J$_{LW}$ throughout this paper, and for convenience
333: express values in
334: units of J$_{21}$, where $J_{21}$ is the mean intensity of the UV background in the LW band in units of
335: $10^{-21}$~erg s$^{-1}$~cm$^{-2}$~Hz$^{-1}$~sr$^{-1}$.
336:
337:
338: The equations of hydrodynamics
339: are solved using the PPM method with a dual energy formulation, which is required
340: to adequately resolve the thermal properties of gas in high-Mach flows. The nonequilibrium
341: chemical
342: evolution and optically thin radiative cooling of the primordial gas is
343: modeled as described in Section~\ref{enzocode}, following 9
344: separate species including molecular hydrogen (but excluding deuterium), with an initial
345: electron fraction of $2.35 \times 10^{-4}$ (which is roughly consistent with
346: \markcite{1968ApJ...153....1P}{Peebles} (1968) for an $\Omega_m = 0.3$, $\Omega_b = 0.04$
347: universe). Note that the initial electron fraction in the
348: calculation is relatively unimportant to the molecular hydrogen formation rates
349: in halo cores, as the electron fraction at the center of a given halo is
350: controlled primarily
351: by mergers and the shock formed by accretion of gas onto the halo.
352:
353: Adaptive
354: mesh refinement is used throughout the innermost high resolution region
355: such that cells are refined by factors of two along each
356: axis, with a maximum of 22 total levels of refinement. This corresponds to a
357: maximum spatial resolution of 115~h$^{-1}$ AU (comoving)
358: at the finest level of resolution, with an overall spatial dynamical range of
359: $5.37 \times 10^8$. To avoid effects due to the finite size of the dark matter
360: particles, the dark matter density is smoothed on a comoving scale of $\sim 0.5$~pc
361: (which corresponds to $\simeq$ 0.03 proper pc at $z \simeq 18$).
362: This is reasonable because at that scale in all of our calculations the gravitational
363: potential in the halo of interest is completely dominated by the baryons.
364:
365: Grid cells are adaptively refined based upon several criteria. We refine on
366: baryon and dark matter overdensities in cells of 4.0 and 8.0, respectively.
367: This corresponds to a maximum mass of gas or dark matter per cell (on the
368: most highly refined static grid) of M$_{max} = 12.78$ and $166.16$~M$_\odot$,
369: respectively. In addition, the \emph{MinimumMassForRefinementLevelExponent}
370: parameter is set to $-0.2$ for both the dark matter and baryon overdensity
371: refinement criteria, meaning that the mass required to refine to a higher
372: level decreases as a function of increasing level, as:
373:
374: \begin{equation}
375: M_{ref}(L) = M_{max} \times 2^{-0.2 L}
376: \label{eqn-refine}
377: \end{equation}
378:
379: where L is the current level of refinement. The negative exponent means
380: that the mass resolution in the calculations is
381: super-Lagrangian -- for example,
382: M$_{ref}(L=20) = 0.0625$~M$_{ref}(L=0)$.
383: In addition to refining on baryon and dark matter overdensity, these simulations include
384: refinement criteria which ensure that shocks are always well-resolved, that the
385: cooling time in a given cell is always longer
386: than the sound crossing time of that cell, and that the Jeans length is always
387: resolved by at least 16 cells. This last criterion guarantees that the Truelove
388: criterion \markcite{truelove97}({Truelove} {et~al.} 1997) is always resolved by a factor of four more cells
389: in each dimension than is strictly necessary, ensuring that no artificial fragmentation
390: will take place.
391:
392: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
393: \section{Mean halo properties at collapse}\label{results.meanprop}
394:
395: Figures~\ref{fig.image_jlw_0} and~\ref{fig.image_jlw_1em21} show projections of
396: baryon density and temperature for the J$_{21} = 0$ and J$_{21} = 1$
397: calculations, respectively, at the epoch at which each calculation collapses,
398: defined as the redshift
399: at which the baryon number density reaches $\simeq 10^{10}$~cm$^{-3}$.
400: Note that due to the rapid evolution of gas at high density, the
401: ``collapse redshift'' depends very weakly on the exact choice of
402: density threshold.
403: The calculations are started from the same set of initial conditions, and
404: the J$_{21} = 1$ calculation is clearly a later stage in the evolution
405: of the J$_{21} = 0$ run. The satellite halos surrounding the primary halo which
406: collapses at z=24.12 in the J$_{21} = 0$ run have merged by z=17.32 when the halo in the J$_{21} = 1$ case collapses (cf. Fig. \ref{fig.merger}). With virial masses of $5.68 \times 10^5 \Ms$ and $1.26 \times 10^7 \Ms$, respectively, the latter halo is more than 20 times as massive as the former. Though quite different in mass, both halos exhibit
407: similar morphologies -- they are extremely centrally-concentrated, and only a single
408: condensed object (that is to say, a primordial protostellar cloud core) is visible in the
409: highest-resolution panel in each image. The existence of a single cloud core
410: is common to all calculations discussed in this paper.
411:
412: \clearpage
413: %%%%%%%%%%%%%%%%%%% FIGURE %%%%%%%%%%%%%%
414: \begin{figure}
415: \begin{center}
416: %%% set textwidth to 0.45 for apj, 0.9 for one-column MS format
417: \includegraphics[width=0.9\textwidth]{f1.eps}
418: \end{center}
419: \caption{
420: Projections of baryon density and temperature at the final output of the J$_{21} = 0$ simulation
421: ($z = 24.119$; M$_{vir} = 5.68 \times 10^{5}$~M$_\odot$).
422: Top row: Projected log baryon density. Bottom row: Projected, mass-weighted log baryon temperature.
423: Left column: Region 2.13 kpc (proper) across and deep.
424: Middle column: Region 133.3 pc (proper) across and deep.
425: Right column: Region 0.130 pc (proper) across and deep.
426: The middle and right column are zoomed images focusing on the high baryon density core which forms in the center of the halo.
427: }
428: \label{fig.image_jlw_0}
429: \end{figure}
430: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
431:
432: %%%%%%%%%%%%%%%%%%% FIGURE %%%%%%%%%%%%%%
433: \begin{figure}
434: \begin{center}
435: \includegraphics[width=0.9\textwidth]{f2.eps}
436: \end{center}
437: \caption{
438: Projections of baryon density and temperature at the final output of the J$_{21} = 1$ simulation
439: ($z = 17.322$; M$_{vir} = 1.260 \times 10^{7}$~M$_\odot$).
440: Top row: Projected log baryon density. Bottom row: Projected, mass-weighted log baryon temperature.
441: Left column: Region 2.13 kpc (proper) across and deep.
442: Middle column: Region 533 pc (proper) across and deep.
443: Right column: Region 0.260 pc (proper) across and deep.
444: The middle and right column are zoomed images focusing on the high baryon density core which forms in the center of the halo.
445: }
446: \label{fig.image_jlw_1em21}
447: \end{figure}
448: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
449: \clearpage
450:
451: Figure~\ref{fig.halovals1} shows several mean halo quantities for all simulations
452: discussed in this paper, including halo collapse redshift as a function
453: of J$_{LW}$, virial mass as a function of J$_{LW}$, virial temperature as a function
454: of J$_{LW}$, and the virial mass as a function of collapse redshift. Simulations where the
455: soft UV background is turned on are shown by solid squares, and the ``control'' J$_{21} = 0$
456: calculation is shown by an open square. In plots where J$_{LW}$ is shown on the x-axis, the
457: J$_{21} = 0$ simulation is placed at log$_{10} J_{LW} = -24.5$.
458:
459: Figure~\ref{fig.halovals1} shows that there is a clear relationship between the LW
460: intensity and the collapse redshift and virial mass of the halo. A larger LW
461: intensity results in a later collapse time and larger virial mass because the halo must be hotter
462: in order to have a cooling time which is less than a Hubble time despite the depressed \h2 abundance in the high-density gas at the halo's center.
463: This is discussed further in Section~\ref{results.evolution} The final mass of the halos in the simulations with
464: J$_{21} = 1$ is approximately a factor of 20
465: higher than that in the ``control'' simulation. This is in qualitative agreement with \markcite{2001ApJ...548..509M}{Machacek} {et~al.} (2001),
466: who suggest that there is a ``minimum halo mass'' as a function of the strength of the LW background of the form
467:
468: \begin{equation}
469: M_{TH}(M_\odot) = 1.25 \times 10^5 + 8.7 \times 10^5 ~\left( \frac{F_{LW}}{10^{-21}} \right)^{0.47}, F_{LW} \leq 10^{-21}
470: \label{eqn-mach-mthresh}
471: \end{equation}
472:
473: This threshold mass is plotted in panel (c) of Figure~\ref{fig.halovals1}.
474: Eq. \ref{eqn-mach-mthresh} is only strictly valid over the range $0 \leq F_{LW}=4\pi J_{LW} \leq 10^{-21}$, because this was the range simulated by \markcite{2001ApJ...548..509M}{Machacek} {et~al.} (2001). We see that our points for $J_{21} \leq 0.1$ parallel the threshold curve but at a mass approximately four times higher. This difference can be ascribed to two factors. First,
475: the threshold mass is the lowest possible halo mass that can collapse, derived from a statistical sample. Since we study only one realization which focuses on the most massive halo in the box, its mass is bound to be higher than the statistical minimum. Second, \markcite{2001ApJ...548..509M}{Machacek} {et~al.} (2001)'s criterion for cooling catches halos at an earlier stage of evolution compared to our data points, which give the halos' virial masses at the time of
476: central baryon collapse (the ``collapse redshift'').
477:
478: Examination of panels (b) and (c) of Figure~\ref{fig.halovals1} show that there is some sort
479: of ``phase change'' between J$_{21} = 10^{-1.5}$
480: and $10^{-1}$ causing the halo mass at the time of collapse to increase
481: steeply and non-monotonically.
482: The collapse redshift steadily
483: decreases as a function of UV increasing background strength
484: (i.e. collapse of gas at the center of the halo is delayed).
485: However, panel (c) shows that the virial mass steadily increases with increasing
486: UV background strength until J$_{21} = 10^{-1.5}$, at which point there is a jump in
487: mass by more than a factor of four, above which the halo mass increases only
488: slightly with increasing UV background strength. That something interesting should happen at these LW intensities is consistent with the findings of~\markcite{2003ApJ...592..645Y}{Yoshida} {et~al.} (2003) who found that \h2 cooling is strongly suppressed at J$_{21}=0.1$. They predicted that star formation would be inhibited since the equilibrium \h2 fraction is below the critical fraction for cooling derived by \markcite{1997ApJ...474....1T}{Tegmark} {et~al.} (1997). Contrary to these predictions, we find collapse not only at J$_{21}=0.1$, but also at values as large as J$_{21}=1$. We note that highest value
489: considered by \markcite{2001ApJ...548..509M}{Machacek} {et~al.} (2001) was J$_{21}= 0.0796$, and the highest simulated by~\markcite{2003ApJ...592..645Y}{Yoshida} {et~al.} (2003) was J$_{21}=0.1$. Therefore our cases J$_{21} > 0.1$ have not been examined before, and certainly not at the resolution of our simulations.
490:
491: How is the gas in the center of these halos able to cool and collapse in such high radiation backgrounds? This is analyzed in some detail in the next section. A hint is provided in (d) of Figure~\ref{fig.halovals1}, which plots the halo virial temperature as a function of J$_{LW}$. At background strengths above
492: J$_{21} = 10^{-1.5}$, the virial temperature is consistently approximately $10^4$~K,
493: which is roughly the temperature at which atomic hydrogen cooling is effective. However, an examination of the radial temperature profiles shows that high-density gas in the center of the halo never reaches these temperatures, but is more typically 2000 K. Atomic line cooling is unimportant at such low temperatures; however the \h2 cooling rate per particle is roughly 100 times as large at 2000K as at 500K. \h2 cooling still operates in the halo centers despite low \h2 abundance due to the higher cooling rates and long evolutionary timescales. In the rest of this section we merely present additional mean properties of the halo and its central region at collapse, and defer discussion of the relevant timescales to Section~\ref{results.evolution}.
494:
495: \clearpage
496: %%%%%%%%%%%%%%%%%%% FIGURE %%%%%%%%%%%%%%
497: \begin{figure}
498: \begin{center}
499: \includegraphics[width=0.9\textwidth]{f3.eps}
500: \end{center}
501: \caption{
502: Mean halo quantities for several simulations with the same cosmic
503: realization but a range of Lyman-Werner molecular hydrogen
504: photodissociating flux
505: backgrounds.
506: Panel (a): J$_{LW}$ vs. halo collapse redshift.
507: Panel (b): halo virial mass vs. halo collapse redshift
508: Panel (c): halo virial mass vs. J$_{LW}$
509: Panel (d): halo virial temperature vs. J$_{LW}$
510: The J$_{21} = 0$ ``control'' result are shown as an open square
511: (and is at log J$_{LW} = -24.5$ in the panels which
512: are a function of J$_{LW}$).
513: In the bottom left panel, the dashed line corresponds to
514: the fitting function for threshold mass
515: from \markcite{2001ApJ...548..509M}{Machacek} {et~al.} (2001), Eqtn. 8.}
516: \label{fig.halovals1}
517: \end{figure}
518: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
519: \clearpage
520:
521: Figure~\ref{fig.halovals2} shows several properties of the halo core at the epoch of
522: collapse, including the core temperature, molecular hydrogen fraction, and spherically-averaged
523: accretion rate as a function of ultraviolet background strength, and the spherically-averaged
524: accretion rate as a function of the molecular hydrogen fraction. All ``core'' values are
525: spherically-averaged and measured at the mass shell where $100$~M$_\odot$ of gas
526: is enclosed. Panel (a) shows that the core H$_2$ fraction decreases significantly with
527: increasing FUV flux, with a corresponding increase in the core temperature (Panel (b)). This
528: relationship is similar to that noted in~\markcite{oshea07a}{O'Shea} \& {Norman} (2007), where the amount of molecular
529: hydrogen at densities of $10^4 - 10^8$~cm$^{-3}$ varies between simulations, and correlates
530: strongly with baryon temperature. Given that the accretion of gas onto the protostellar
531: cloud is subsonic, this results in a strong relationship
532: between the soft UV background flux and accretion rate onto the protostellar cloud, as shown by
533: Panel (c), with accretion rates varying
534: by more than a factor of 30 between the J$_{21} = 0$ and $1$ cases. Panel (d) shows
535: the strength of the correlation between core H$_2$ fraction and accretion rate. Note that
536: the values discussed above are insensitive to the exact definition of the halo ``core.''
537:
538: \clearpage
539: %%%%%%%%%%%%%%%%%%% FIGURE %%%%%%%%%%%%%%
540: \begin{figure}
541: \begin{center}
542: \includegraphics[width=0.9\textwidth]{f4.eps}
543: \end{center}
544: \caption{
545: Mean quantities within the central $100 \Ms$ core for several simulations with the same cosmic
546: realization but a range of LW background intensities.
547: Panel (a): J$_{LW}$ vs. baryon core temperature.
548: Panel (b): J$_{LW}$ vs. baryon core $H_2$ fraction.
549: Panel (c): J$_{LW}$ vs. instantaneous accretion rate.
550: Panel (d): Baryon core $H_2$ fraction vs. accretion rate.
551: The J$_{21} = 0$ ``control'' result are shown as an open square
552: (and is at log J$_{LW} = -24.5$ in the panels which
553: are a function of J$_{LW}$). All ``core'' values are spherically-averaged and
554: measured at the mass shell where
555: $100$~M$_\odot$ of gas is enclosed.
556: }
557: \label{fig.halovals2}
558: \end{figure}
559: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
560: \clearpage
561:
562: Figure~\ref{fig.halovals3} shows the halo core mass and mass fraction at the epoch of collapse
563: (note that the definition of ``core'' is somewhat different than in the previous figure).
564: The halo ``core mass'' is defined as being all gas at a density of n$_H = 10^4$~cm$^{-3}$ or above,
565: and the core mass fraction is defined as the core mass divided by the virial mass of the halo at the epoch
566: of collapse. This choice of minimum density ensures that we have captured the entirety of the quasistatically contracting
567: analog of a galactic molecular cloud core, and for practical purposes is comparable to the gas fraction of
568: cold, dense gas discussed in~\markcite{2001ApJ...548..509M}{Machacek} {et~al.} (2001).
569: The size of the halo core increases in absolute terms as the strength of the photodissociating
570: background is increased, from $\simeq 10^3$~M$_\odot$ until J$_{21} = 0.1$, where it reaches $\simeq 11,000$~M$_\odot$.
571: Above this value of J$_{21}$, the core mass then begins to decrease, reaching $\simeq 7600$~M$_\odot$ at J$_{21} = 1$.
572: The core mass fraction generally decreases (though with some noise in the relationship) from
573: $\simeq 1.7 \times 10^{-3}$ at J$_{21} = 0$ to $\simeq 6.96 \times 10^{-4}$ at J$_{21} = 1$, though the total
574: overall change is less than a factor of 2.
575:
576: \clearpage
577: %%%%%%%%%%%%%%%%%%% FIGURE %%%%%%%%%%%%%%
578: \begin{figure}
579: \begin{center}
580: \includegraphics[width=0.9\textwidth]{f5.eps}
581: \end{center}
582: \caption{
583: Halo core mass (Panel (a)) and core mass fraction (Panel (b)) for several
584: simulations with the same cosmic
585: realization but a range of LW background intensities.
586: The ``core mass'' is defined as the mass of gas in the halo which has a density
587: of n$_H = 10^4$~cm$^{-3}$ or above. The ``core mass fraction'' is defined as the ``core
588: mass'' divided by the virial mass of the halo at the epoch of collapse.
589: The J$_{LW} = 0$ ``control'' result are shown as an open square
590: (and is at log J$_{LW} = -24.5$ in the panels which
591: are a function of J$_{LW}$).
592: }
593: \label{fig.halovals3}
594: \end{figure}
595: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
596: \clearpage
597:
598: \section{Central evolution prior to collapse}\label{results.evolution}
599:
600: We now examine in more detail the evolution of the conditions in the cloud center
601: leading to collapse. In Figure~\ref{fig.evolution} we plot the central density,
602: temperature, entropy $S \equiv T/n_H^{2/3}$, and \h2 fraction versus time for
603: the calculations with $J_{21}=10^{-3}$ (solid line), $10^{-2}$ (short-dashed line),
604: $10^{-1}$ (long-dashed line) and $0$ (dot-dashed line).
605: In these plots, central values of temperature and density are defined as being
606: the values of these quantities in the cell with the highest baryon density.
607: It is most instructive to compare
608: the two extreme cases, J$_{21}=10^{-3}$, which very closely resembles the
609: J$_{21}=0$ case, and the J$_{21}=1$ case, which according to~\markcite{2003ApJ...592..645Y}{Yoshida} {et~al.} (2003)
610: should not collapse at all. In panel (a) we see the central density increase modestly
611: from $\sim$ 2 cm$^{-3}$ to $\sim$ 20 cm$^{-3}$ over 40 Myr, and then increase rapidly
612: thereafter, reaching $10^8$ cm$^{-3}$ in a scant additional 15 Myr. This evolution is
613: driven by \h2 cooling, as can be seen from the entropy evolution in panel (c). As
614: expected, runaway cooling occurs when the \h2 fraction reaches $\sim 2 \times 10^{-4}$
615: (panel d), consistent with the \markcite{1997ApJ...474....1T}{Tegmark} {et~al.} (1997) analysis. By contrast the
616: J$_{21}=1$ case requires 120 Myr for the central density to increases from $\sim$ 2
617: cm$^{-3}$ to $\sim$ 20 cm$^{-3}$. As shown in panel (c), the first 50 Myr of this
618: contraction is adiabatic, followed by an increase in entropy due to some heating
619: event (mergers). The fact that central entropy is not decreasing for 90 Myr is indicative
620: of the fact that \h2 cooling is unimportant over this interval due to extremely low
621: equilibrium \h2 fractions (f$_{H2} < 10^{-6}$; (panel d)).
622:
623: However, something interesting happens in the J$_{21} = 1$ calculation
624: at $t \approx 170$ Myr. Prior to that time the
625: central temperature has crept up to 5000 K due to the increased virial mass. The cooling
626: rate per \h2 molecule is about 3 orders of magnitude higher at 5000K than at 500K -- a
627: typical temperature at the halo center in low UV background evolutions. We believe this high temperature
628: is due to a combination of adiabatic heating as the potential well deepens, and merger-induced
629: shock heating. As shown in Figure \ref{fig.merger}, three large halos merge between $z=20$ and $z=18$,
630: quadrupling the halo mass (Figure \ref{fig.halovals1}, panel (b)). The elevated cooling rates
631: cool the gas at the halo center to $T \approx 2000K$, allowing the gas density to increase slightly.
632: The two effects reduce the central entropy from $S \approx$ 700 K cm$^2$ to $\approx$ 250 K cm$^2$.
633: This is followed by another heating event at t=190 Myr, presumably due to another merger,
634: followed by a second more catastrophic cooling event. At t=200 Myr, the halo core collapses,
635: driving the central density higher, temperature and entropy lower, and \h2 fraction higher.
636:
637: Why does this collapse occur? This is analyzed in Figure \ref{fig.timescales}, where we plot
638: the evolution of the important timescales in the center of the halo. In the J$_{21}=10^{-3}$
639: case, the 2-body \h2 formation timescale is always roughly an order of magnitude smaller than the
640: photodissociation timescale. Consequently, the \h2 abundance is out of equilibrium and grows
641: steadily with time (Figure \ref{fig.evolution} panel d). The \h2 cooling time in the central
642: zone ($\leq 1$ Myr) is far less than the Hubble time ($\sim 100$ Myr), and before too long
643: runaway collapse occurs. At higher values of J$_{LW}$, the 2-body \h2 formation time always
644: hovers around the photodissociation timescale, regardless of how long that timescale is. This tells
645: us that \h2 is in equilibrium: the 2-body formations balance the photo-destructions by the
646: LW background. The equilibrium \h2 fraction is given by
647:
648: \begin{equation}
649: f_{H2} \approx \frac{k_{H^-}n_e}{k_{diss}}
650: \label{eqn-fh2}
651: \end{equation}
652:
653: where k$_{H^-}$ is the rate coefficient for the formation of H$^-$ (the limiting reaction
654: for H$_2$ formation at the temperature and density range considered here), $n_e$ is the
655: electron density, and k$_{diss}=1.1 \times 10^8 F_{LW} s^{-1}$ is
656: the \h2 photodissociation rate due to the Solomon process~\markcite{abel97,2003ApJ...592..645Y}({Abel} {et~al.} 1997; {Yoshida} {et~al.} 2003).
657: Using T $=2000$~K and $n_e=10^{-4}$, we get $f_{H2}=1.46\times
658: 10^{-6}~\left( \frac{F_{LW}}{10^{-21}} \right)^{-1}$, which is in good agreement with
659: the \h2 fractions we see in Figure \ref{fig.evolution}. Since
660: $t_{H2}\equiv n_{H2}$/ \.{n}$_{H2} \propto f_{H2}$, the smaller $f_{H2}$ is, the shorter
661: its formation time.
662:
663: Now let us consider the \h2 cooling time evolution (short-long-dashed line). It is far lass than a
664: Hubble time for J$_{21} \leq 10^{-1}$, ensuring that these halo cores will eventually
665: cool and collapse. In the J$_{21}=1$ case, the cooling time drops below the Hubble time for the
666: first time at t=170 Myr, and again at t=190 Myr where it begins a steady decrease toward
667: $10^6$ yr. Despite the low equilibrium \h2 fraction, if we wait long enough the inexorable
668: press of time eventually establishes the condition $t_{cool} < t_{Hubble}$ and we get
669: runaway collapse, even for J$_{21}=1$. Note that we get catastrophic cooling because we are
670: in the density regime ($n<10^4$ cm$^{-3}$) where \h2 cooling is proportional to the square
671: of the gas density, not linear in the gas density~\markcite{2000ApJ...540...39A,ABN02}({Abel}, {Bryan}, \& {Norman} 2000; {Abel} {et~al.} 2002).
672:
673: \clearpage
674: %%%%%%%%%%%%%%%%%%% FIGURE %%%%%%%%%%%%%%
675: \begin{figure}
676: \begin{center}
677: \includegraphics[width=0.9\textwidth]{f6.eps}
678: \end{center}
679: \caption{
680: Evolution of the central density (Panel (a)),
681: temperature (Panel (b)), entropy (Panel (c)),
682: and \h2 fraction (Panel (d))as a function
683: of time for the cases $J_{21}=10^{-3}$ (solid), $10^{-2}$ (short-dashed), $10^{-1}$
684: (long-dashed) and $1$ (dot-dashed).
685: }
686: \label{fig.evolution}
687: \end{figure}
688: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
689:
690: %%%%%%%%%%%%%%%%%%% FIGURE %%%%%%%%%%%%%%
691: \begin{figure}
692: \begin{center}
693: \includegraphics[width=0.9\textwidth]{f7.eps}
694: \end{center}
695: \caption{
696: Evolution of timescales in the central zone for the cases $J_{21}=10^{-3}, 10^{-2},
697: 10^{-1}$ and $1$ (Panels (a)--(d), respectively):
698: 2-body \h2 formation time (solid); 3-body \h2 formation time (short-dashed);
699: \h2 collisional dissociate time (long-dashed);
700: \h2 photodissociation time (dot-dashed);
701: \h2 cooling
702: time (short-long-dashed).
703: }
704: \label{fig.timescales}
705: \end{figure}
706: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
707:
708: %%%%%%%%%%%%%%%%%%% FIGURE %%%%%%%%%%%%%%
709: \begin{figure}
710: \begin{center}
711: \includegraphics[width=0.9\textwidth]{f8.eps}
712: \end{center}
713: \caption{
714: Merger history of the $M_{vir}=1.26 \times 10^7 \Ms$, T$_{vir}=10^4$~K halo that collapses
715: at z=17.32 with a Lyman-Werner background flux $J_{21}=1$. In comic strip order, the redshifts are
716: z=22, 21, 20, 19, 18, and 17.32. Field of view is 53.571 comoving kpc (2.678 proper
717: kpc at z = 19). Logarithm of the projected baryon density (column density) is displayed, with color table scaled
718: to the maximum and minimum values in each image.
719: }
720: \label{fig.merger}
721: \end{figure}
722: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
723: \clearpage
724:
725:
726: \section{Halo properties at the epoch of collapse}\label{results.collapse}
727:
728: Figure~\ref{fig.radprof_zcoll.1} shows radial profiles of baryon number density, temperature,
729: enclosed mass, circular velocity, and RMS Mach number as a function of radius, as well as
730: specific angular momentum as a function of enclosed mass, for all simulations discussed in
731: this paper. To facilitate comparison, results from the output of each simulation where the peak
732: baryon number density at the center of the halo is approximately $10^{10}$ cm$^{-3}$ are shown, in
733: order to capture each halo at a similar evolutionary stage, rather than at a fixed
734: point in time. Panel (a) shows that all of the simulations have similar density profiles.
735: The scatter in number density at a given radius is readily explained by variation in halo
736: mass -- halos which collapse at later times (due to higher UV background strength) are more
737: massive, and thus have higher overall baryon densities. This is shown in another way in
738: Panel (c), in a plot of enclosed mass as a function of radius which shows that all halos
739: have very similar profiles at $r \la 10^{-2}$ pc, but a significant variation at larger
740: radii which is related to halo mass. The plot of temperature as a function of radius in
741: Panel (b) shows an interesting trend -- as the LW background is increased, the overall halo
742: temperature as well as the halo core (where ``core'' is roughly defined as gas within
743: $\sim 1$ pc of the halo center) temperature go up. In the outskirts of the halo, where
744: baryon densities are low and thus cooling times are long, this is due primarily to the
745: increase in halo mass, with the peak temperature corresponding approximately to the halo
746: virial temperature. In the halo core, however, this temperature is correlated more strongly
747: with the H$_2$ fraction, and thus the UV background, as suggested by
748: Figure~\ref{fig.halovals2}. The plots of specific angular momentum as a function of enclosed
749: mass and Keplerian velocity fraction (defined as the local circular
750: velocity divided by the Keplerian orbital velocity due to mass within that radius)
751: as a function of radius in panels (d) and (e) show some mild trends
752: trends. It appears that halos in simulations with stronger UV backgrounds tend to have
753: less specific angular momentum at a given mass shell, and also tend to have a lower
754: Keplerian velocity fraction at a given radius than gas in halos which form in the
755: presence of a lower UV background -- indeed, the halo which forms in the J$_{21} = 1$
756: calculation has the least angular momentum of all of the simulations, and the lowest
757: Keplerian velocity fraction out to $\sim 1$~pc. We speculate that this is due to
758: transport and segregation of angular momentum by turbulence -- halos whose collapse
759: is delayed have more time for these processes to act within the halo core, resulting
760: in less angular momentum overall when the halo ultimately collapses (note that this will
761: be discussed more fully in a later paper).
762: The plot of RMS Mach number as a function of radius
763: in panel (f) also
764: shows no obvious trend, though at radii below $\sim 10^{-2}$ pc, simulations with a smaller
765: UV background generally have higher RMS Mach numbers. This is due primarily to the gas
766: being colder, and not the gas velocities being higher.
767:
768: \clearpage
769: %%%%%%%%%%%%%%%%%%% FIGURE %%%%%%%%%%%%%%
770: \begin{figure}
771: \begin{center}
772: \includegraphics[width=0.3\textwidth]{f9a.eps}
773: \includegraphics[width=0.3\textwidth]{f9b.eps}
774: \includegraphics[width=0.3\textwidth]{f9c.eps}
775: \includegraphics[width=0.3\textwidth]{f9d.eps}
776: \includegraphics[width=0.3\textwidth]{f9e.eps}
777: \includegraphics[width=0.3\textwidth]{f9f.eps}
778: \end{center}
779: \caption{
780: Evolution of several spherically-averaged baryon quantities for simulations
781: with all values of J$_{21}$.
782: Panel (a): $\rho_b r^2$ as a function of radius.
783: Panel (b): Temperature as a function of radius.
784: Panel (c): Enclosed gas mass as a function of radius.
785: Panel (d): Specific angular momentum as a function of enclosed mass.
786: Panel (e): Keplerian velocity fraction as a function of radius.
787: Panel (f): Gas RMS Mach number as a function of radius.
788: All quantities except enclosed gas mass are mass-weighted, and
789: all simulations are shown at the point where the maximum
790: number density is approximately $10^{10}$~cm$^{-3}$.
791: Line types and weights correspond to simulations, as follows.
792: Black solid line: J$_{21} = 0$.
793: Black short-dashed line: J$_{21} = 10^{-3}$.
794: Black long-dashed line: J$_{21} = 10^{-2.5}$.
795: Black dot short-dashed line: J$_{21} = 10^{-2}$.
796: Black dot long-dashed line: J$_{21} = 10^{-1.5}$.
797: Black short dashed-long dashed line: J$_{21} = 10^{-1}$.
798: Red solid line: J$_{21} = 10^{-0.67}$.
799: Red short-dashed line: J$_{21} = 10^{-0.33}$.
800: Red long-dashed line: J$_{21} = 1$.
801: }
802: \label{fig.radprof_zcoll.1}
803: \end{figure}
804: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
805: \clearpage
806:
807: Figure~\ref{fig.radprof_zcoll.2} shows the H$_2$ fraction, electron fraction, H$^-$ fraction,
808: ratio of cooling time to sound crossing time, ratio of cooling time to dynamical time, and ratio
809: of sound crossing time to dynamical time as a function of radius, for all
810: simulations. Outputs and line types correspond to those in Figure~\ref{fig.radprof_zcoll.1}.
811: There is a clear relationship at all radii between H$_2$ fraction and UV background strength
812: in panel (a) -- as
813: the FUV background is increased, the overall H$_2$ fraction decreases. This difference
814: is most noticeable at radii of $0.1 - 10$ pc, but is maintained at larger and smaller radii.
815: There is a ``kink'' at approximately $10^{-2}$ pc where the values are all quite similar. This
816: corresponds to a baryon number density of $\sim 10^8$~cm$^{-3}$ in all simulations, which is
817: where 3-body H$_2$ formation begins to occur. Panels (b) and (c) show that
818: e$^-$ and $H^-$ fractions track each other,
819: which is to be expected -- the local electron fraction controls the amount of H$^-$ which
820: can be produced, which is the limiting reaction in the dominant mode of H$_2$ formation for
821: n $\la 10^8$ cm$^{-3}$ -- and there is a general trend with increasing electron and H$^-$ fractions with
822: increasing UV background flux.
823:
824: The plots of the ratio of cooling time to sound crossing time, ratio of cooling time to
825: dynamical time, and ratio of sound crossing time to dynamical time as a function of radius
826: shown in panels (d), (e) and (f)
827: display some interesting trends. The three simulations with the lowest UV background strengths
828: have a much longer cooling time than the other calculations, which are all grouped roughly
829: together with no discernible pattern. This agrees well with the plot of the ratio of cooling
830: time to dynamical time, where a similar trend is observed. The difference between the three
831: simulations with the lowest UV background strengths and the others is due to the somewhat higher
832: H$_2$ fraction in the halo core of these calculations. The gas temperature is $\sim 200$~K at
833: the center of the halo cores in these calculations, and the gas cannot cool further, which results
834: in a very long cooling time. The sound crossing time is also increased, but not as significantly.
835: All simulations have approximately the same ratio of sound crossing to dynamical times at all radii.
836: Given that the cooling time is typically longer than both the sound crossing time and dynamical time,
837: one can infer that the collapse of the halo is occurring quasi-statically for all simulations.
838:
839: \clearpage
840: %%%%%%%%%%%%%%%%%%% FIGURE %%%%%%%%%%%%%%
841: \begin{figure}
842: \begin{center}
843: \includegraphics[width=0.3\textwidth]{f10a.eps}
844: \includegraphics[width=0.3\textwidth]{f10b.eps}
845: \includegraphics[width=0.3\textwidth]{f10c.eps}
846: \includegraphics[width=0.3\textwidth]{f10d.eps}
847: \includegraphics[width=0.3\textwidth]{f10e.eps}
848: \includegraphics[width=0.3\textwidth]{f10f.eps}
849: \end{center}
850: \caption{
851: Several spherically-averaged baryon quantities for simulations
852: with all values of J$_{21}$.
853: Panel (a): Molecular hydrogen fraction as a function of radius.
854: Panel (b): Electron fraction as a function of radius.
855: Panel (c): $H^-$ fraction as a function of radius.
856: Panel (d): Ratio of gas cooling time to sound crossing time as a function of radius.
857: Panel (e): Ratio of gas cooling time to dynamical time as a function of radius.
858: Panel (f): Ratio of gas sound crossing time to dynamical time as a function of radius.
859: All quantities except enclosed gas mass are mass-weighted, and
860: all simulations are shown at the point where the maximum
861: number density is approximately $10^{10}$~cm$^{-3}$.
862: Line types and weights correspond to those in Figure~\ref{fig.radprof_zcoll.1}.
863: }
864: \label{fig.radprof_zcoll.2}
865: \end{figure}
866: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
867: \clearpage
868:
869: Figure~\ref{fig.radprof_zcoll.3} shows radial velocity as a function of radius,
870: accretion time (defined as $M_{b,enc}/\dot{M}$) as a function of enclosed baryon mass
871: $\dot{M}$ as a function of enclosed baryon mass and $\dot{M}$ as a function of time
872: for all simulations. Output time and lines correspond to those in
873: Figures~\ref{fig.radprof_zcoll.1} and~\ref{fig.radprof_zcoll.2}.
874: Panel (a) shows a clear relationship between the strength of the soft UV background
875: and the radial velocity, with the calculations that have higher UV backgrounds
876: also having higher infall velocities. This is easily understood by the quasistatic
877: contraction of the gas, which takes place at or below the local sound speed in the gas,
878: as discussed by \markcite{ABN02}{Abel} {et~al.} (2002) and \markcite{oshea07a}{O'Shea} \& {Norman} (2007). Given that the halo temperatures
879: are systematically higher in calculations where the UV background is stronger, the sound
880: speed is higher, and thus the rate at which the gas at the center of the halo contracts is higher. This leads to
881: lower accretion times (Panel (b)) and higher accretion rates (Panel (c)), with the accretion
882: rates for the simulations with the strongest UV backgrounds being higher than those with
883: a low (or no) UV background. Panel (d) shows that the time evolution of the accretion
884: rate in all simulations is qualitatively similar, but varies strongly in absolute magnitude
885: and in the time at which the accretion rate onto the evolving protostellar cloud peaks,
886: suggesting that the stars forming in these simulations may have very different evolutionary
887: histories.
888:
889: \clearpage
890: %%%%%%%%%%%%%%%%%%% FIGURE %%%%%%%%%%%%%%
891: \begin{figure}
892: \begin{center}
893: \includegraphics[width=0.45\textwidth]{f11a.eps}
894: \includegraphics[width=0.45\textwidth]{f11b.eps}
895: \includegraphics[width=0.45\textwidth]{f11c.eps}
896: \includegraphics[width=0.45\textwidth]{f11d.eps}
897: \end{center}
898: \caption{
899: Several spherically-averaged baryon quantities for simulations
900: with all values of J$_{21}$.
901: Panel (a): Radial velocity as a function of radius.
902: Panel (b): Accretion time as a function of enclosed baryon mass.
903: Panel (c): Instantaneous accretion rate as a function of enclosed baryon mass.
904: Panel (d): Estimated accretion rate as a function of time.
905: All quantities except enclosed gas mass are mass-weighted, and
906: all simulations are shown at the point where the maximum
907: number density is approximately $10^{10}$~cm$^{-3}$.
908: Line types and weights correspond to those in Figure~\ref{fig.radprof_zcoll.1}.
909: The upper and lower light short-long-dashed curves which extend from the
910: upper left corner of Panel (b) correspond to
911: the main sequence lifetime of a massive Population III star of that mass and
912: the Kelvin-Helmholtz timescale of a Population III timescale with a given
913: luminosity and
914: radius. All values are taken from~\markcite{2002A&A...382...28S}{Schaerer} (2002).
915: The three light diagonal short-dashed lines which extend
916: from bottom left to top right in panel (b) correspond to masses
917: accreted using constant accretion rates of (from top to bottom)
918: $\dot{m} = 10^{-3}$, $10^{-2}$, and $10^{-1}$ M$_\odot/$yr.
919: The light horizontal short-dashed line in panels (c) and
920: (d) correspond to the ``critical'' accretion rate of~\markcite{2003ApJ...589..677O}{Omukai} \& {Palla} (2003),
921: $\dot{m} \simeq 4 \times 10^{-3}$~M$_\odot/$yr.
922: }
923: \label{fig.radprof_zcoll.3}
924: \end{figure}
925: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
926: \clearpage
927:
928: \section{Halo properties at fixed redshift}\label{results.fixred}
929:
930: Figure~\ref{fig.radprof_fixred.1} shows the status of several
931: spherically-averaged quantities for halos from all simulations at a fixed
932: redshift, $z=25$, which is shortly before the baryons in the most massive halo
933: in the J$_{21} = 0$ simulation collapses to high
934: density. This is at an early point in the evolution of the majority of these halos, and
935: several conclusions can be drawn from the plots of number density, temperature, H$_2$ fraction,
936: and radial velocity as a function of radius. The molecular hydrogen fraction declines monotonically
937: with increasing UV background strength, in agreement with an estimate for equilibrium H$_2$ fraction at low
938: ($n \ll 10^8$~cm$^{-3}$) densities:
939:
940: \begin{equation}
941: f_{H2} = \frac{k_{H^-}n_e}{k_{LW}}
942: \label{eqn-fh2}
943: \end{equation}
944:
945: where k$_{H^-}$ is the rate coefficient for the formation of H$^-$ (the limiting reaction
946: for H$_2$ formation at the temperature and density range considered here) and k$_{LW}$ is
947: the rate coefficient for H$_2$ in the LW band~\markcite{abel97,2003ApJ...592..645Y}({Abel} {et~al.} 1997; {Yoshida} {et~al.} 2003).
948: Halo cores at higher central densities are departing from the approximations used in the above
949: estimate and thus have values of the H$_2$ fraction that are not quite the equilibrium
950: values. Halo cores which have reached core H$_2$ fractions higher than $\sim 10^{-4}$
951: have been able to cool to significantly below the virial temperature (since the cooling
952: time is much less than the Hubble time at that value of the H$_2$ fraction), allowing
953: the baryon density to increase. The effect of cooling can also be seen in the plot of
954: radial velocity as a function of radius, where the core regions of halos that have cooled are
955: contracting (albeit very slowly), while those which are still at the virial temperature
956: are not showing significant signs of contraction within $\sim 20$ pc. All halos have
957: some inflow at large radii, due to the infall of gas from adjoining filaments.
958:
959: \clearpage
960: %%%%%%%%%%%%%%%%%%% FIGURE %%%%%%%%%%%%%%
961: \begin{figure}
962: \begin{center}
963: \includegraphics[width=0.45\textwidth]{f12a.eps}
964: \includegraphics[width=0.45\textwidth]{f12b.eps}
965: \includegraphics[width=0.45\textwidth]{f12c.eps}
966: \includegraphics[width=0.45\textwidth]{f12d.eps}
967: \end{center}
968: \caption{
969: Several spherically-averaged baryon quantities for simulations
970: with all values of J$_{21}$.
971: Panel (a): Number density as a function of radius.
972: Panel (b): Temperature as a function of radius.
973: Panel (c): $H_2$ fraction as a function of radius.
974: Panel (d): Radial velocity as a function of radius.
975: All quantities except enclosed gas mass are mass-weighted, and
976: all simulations are shown at $z=25$ (shortly before the J$_{21} = 0$
977: simulation collapses to high density.
978: Line types and weights correspond to simulations, as follows.
979: Black solid line: J$_{21} = 0.0$.
980: Black short-dashed line: J$_{21} = 10^{-3}$.
981: Black long-dashed line: J$_{21} = 10^{-2.5}$.
982: Black dot short-dashed line: J$_{21} = 10^{-2}$.
983: Black dot long-dashed line: J$_{21} = 10^{-1.5}$.
984: Black short dashed-long dashed line: J$_{21} = 10^{-1}$.
985: Red solid line: J$_{21} = 10^{-0.67}$.
986: Red short-dashed line: J$_{21} = 10^{-0.33}$.
987: Red long-dashed line: J$_{21} = 1$.
988: }
989: \label{fig.radprof_fixred.1}
990: \end{figure}
991: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
992: \clearpage
993:
994: \section{Evolution of two representative halos}\label{results.rephalos}
995:
996: In this section we compare the evolution of two representative simulations.
997: We choose the calculations which correspond to J$_{21} = 0$ and $1$.
998: These two simulations represent
999: the extremes in halo evolution -- the J$_{21} = 0$ calculation is an
1000: example of the evolution of a ``standard'' halo, as discussed in previous
1001: literature, and the J$_{21} = 1$ simulation is the most extreme example
1002: of a Population III protostar forming in a T$_{vir} \sim 10^4$~K halo
1003: available from our suite of calculations. As described in Section~\ref{sect.simsetup},
1004: both calculations start from the same initial conditions, with only the strength of
1005: the LW background being different.
1006:
1007: Projections of the log baryon density and temperature for the J$_{21} = 0$ and $1$
1008: simulations are shown in Figures~\ref{fig.image_jlw_0} and~\ref{fig.image_jlw_1em21},
1009: respectively. These projections are at the epoch of collapse, which is $z=24.1$~$(17.3)$
1010: for the J$_{21} = 0$~$(1)$ simulation. The halo masses are significantly
1011: different: $5.68 \times 10^5$ and $1.26 \times 10^7$~M$_\odot$ for the two calculations,
1012: reflecting the rapid pace of mergers at that redshift, and the amount of structure evolution
1013: that takes place over that relatively short time period ($\sim 8.2 \times 10^7$ years). This
1014: can clearly be seen in the left column of both figures, when the large
1015: amount of structure apparent in Figure~\ref{fig.image_jlw_0} has merged into the main
1016: halo by the time of protostellar cloud formation in the J$_{21} = 1$ case, as seen
1017: in Figure~\ref{fig.image_jlw_1em21}.
1018:
1019: Despite the differences in evolutionary states, the halos themselves are quite similar in
1020: Figures~\ref{fig.image_jlw_0} and~\ref{fig.image_jlw_1em21}. Both calculations show halos
1021: that are strongly centrally concentrated, as shown by the center column in both figures.
1022: This column shows projections of the halos which are centered on the baryon peak and are
1023: approximately one virial radius across (and thus are scaled differently in both images).
1024: No fragmentation of the halo core is visible. The rightmost panel in both calculations shows
1025: the center of the halo, where the protostellar cloud is evolving. There is still no evidence for
1026: fragmentation up to a central baryon density of $n \sim 10^{10}$~cm$^{-3}$.
1027:
1028: Figures~\ref{fig.radprof_jlw0.1} and~\ref{fig.radprof_jlw1em21.1} show the temporal
1029: evolution of several spherically-averaged baryon quantities for the
1030: J$_{21} = 0$ and $1$ calculations,
1031: respectively. These plots show baryon number density, enclosed baryon mass,
1032: temperature, H$_2$ fraction, ratio of cooling time to sound crossing time, and
1033: ratio of cooling time to dynamical time, as a function of radius, evolving from a
1034: number density of $\simeq 10$~cm$^{-3}$ to $\simeq 10^{10}$~cm$^{-3}$ in both calculations.
1035: Some qualitative commonalities are obvious between the two calculations. Both halos
1036: steadily grow in central density (Panel (a)), which is coupled to the growing molecular
1037: hydrogen fraction in the central region of the halo (Panel (d)). Both halos collapse quasi-statically, as
1038: shown in panels (e) and (f). Also, the temperature profiles (as shown in Panel (c)) are
1039: broadly similar, in that the gas in the halo central region cools to some minimum value at r $\sim 1-2$ pc
1040: (corresponding to n $\sim 10^4$~cm$^{-3}$ in both simulations) and then creeps steadily upward
1041: as gas collapses to higher densities.
1042:
1043: Though there are qualitative similarities between the evolution of the two calculations, detailed
1044: examination shows some significant quantitative differences. The time that the halo in the J$_{21} = 0$
1045: simulation takes to evolve from a central density of $n \simeq 10$~cm$^{-3}$ to
1046: $10^{10}$~cm$^{-3}$ is $\simeq 24$~Myr, with 21 Myr of that being the time required to
1047: evolve from $n \simeq 10$ to n $\simeq 10^4$ cm$^{-3}$. The time required for
1048: the J$_{21} = 1$ calculation
1049: to evolve from $n \simeq 10$~cm$^{-3}$ to $10^{10}$~cm$^{-3}$ is slightly shorter,
1050: $\simeq$ 17 Myrs, with 8 Myrs required to evolve to n $\simeq 10^{4}$~cm$^{-3}$.
1051: The temperature evolution is also somewhat different -- the J$_{21} = 0$ simulation
1052: has a minimum temperature at r $\sim 1$ pc of 200 K, with n $\simeq 10^4$$^{-3}$
1053: and a molecular hydrogen fraction of $\simeq 10^{-3}$, and roughly 1000 M$_\odot$
1054: of enclosed gas. As the gas within this radius evolves to higher densities, the
1055: temperature increases by a factor of a few, peaking at T $\sim 500$ K at
1056: n~$\sim 10^{10}$~cm$^{-3}$, with f$_{H_2} \simeq 0.02$. The gas cooling time
1057: within r $\sim 1$ pc in the J$_{21} = 0$ simulation is always larger than the sound
1058: crossing and dynamical times by at least a factor of two, implying a quasistatic
1059: gas contraction at all times.
1060:
1061: The J$_{21} = 1$ simulation evolves somewhat differently. The collapsing
1062: gas reaches a minimum temperature of $\sim 800$~K at r $\sim 2$ pc, with a baryon
1063: density of $\sim 2 \times 10^3$~cm$^{-3}$, an enclosed gas mass
1064: of roughly $10^4$ M$_\odot$, and a molecular hydrogen fraction of $\sim 10^{-5}$,
1065: two orders of magnitude below that in the J$_{21} = 0$ calculation at the
1066: equivalent temperature minimum. The low H$_2$ fraction is due to the high
1067: LW background radiation flux. As the halo evolves to higher densities, the gas
1068: temperature and H$_2$ fraction also creep upwards. The temperature of the gas when
1069: the peak reaches roughly $10^{10}$~cm$^{-3}$ is $\sim 1200$ K, with a H$_2$ fraction of
1070: $\sim 4 \times 10^{-3}$. The cooling time at r $\sim 2$ pc is an order of magnitude
1071: higher than the sound crossing or dynamical times, implying an extremely slow contraction
1072: of the gas at that radius. As the central density increases, however, the cooling
1073: time becomes lower, with T$_{cool}$/T$_{dyn} \sim 0.7$ and T$_{cool}$/T$_{cross}$
1074: dipping to $\sim 0.8$ at its minimum, but the latter increasing toward the
1075: central density peak to a ratio of $\sim 10$. This implies a more rapid contraction
1076: of gas in the halo center in the simulation with a higher UV background, despite the lack of
1077: molecular hydrogen. It is also worth noticing that the J$_{21} = 1$ calculation
1078: has H$_2$ fractions at n $\sim 10^4$ and $10^8$~cm$^{-3}$ of $\simeq 2 \times 10^{-5}$
1079: and $8 \times 10^{-4}$, respectively, as compared to f$_{H_2} \simeq 10^{-3}$ and
1080: $2 \times 10^{-3}$ at the same densities in the J$_{21} = 0$ calculation.
1081:
1082: \clearpage
1083: %%%%%%%%%%%%%%%%%%% FIGURE %%%%%%%%%%%%%%
1084: \begin{figure}
1085: \begin{center}
1086: \includegraphics[width=0.3\textwidth]{f13a.eps}
1087: \includegraphics[width=0.3\textwidth]{f13b.eps}
1088: \includegraphics[width=0.3\textwidth]{f13c.eps}
1089: \includegraphics[width=0.3\textwidth]{f13d.eps}
1090: \includegraphics[width=0.3\textwidth]{f13e.eps}
1091: \includegraphics[width=0.3\textwidth]{f13f.eps}
1092: \end{center}
1093: \caption{
1094: Evolution of several spherically-averaged baryon quantities as a function
1095: of time for the J$_{21} = 0$ simulation.
1096: Panel (a): Number density as a function of radius.
1097: Panel (b): Enclosed gas mass as a function of radius.
1098: Panel (c): Temperature as a function of radius.
1099: Panel (d): $H_2$ fraction as a function of radius.
1100: Panel (e): Ratio of cooling time to sound crossing time as a function of radius.
1101: Panel (f): Ratio of cooling time to dynamical time as a function of radius.
1102: All quantities except enclosed gas mass are mass-weighted.
1103: Line types and weights correspond to different times, as follows.
1104: Red short dashed-long dashed line: $z=27.678$, $t=1.1088 \times 10^8$ years.
1105: Red dot long-dashed line: $z=26.010$, $\Delta t = 1.0428 \times 10^7$ years.
1106: Red dot short-dashed line: $z = 24.565$, $\Delta t = 1.0428 \times 10^7$ years.
1107: Red long-dashed line: $z = 24.273$, $\Delta t = 2.291 \times 10^6$ years.
1108: Red short-dashed line: $z = 24.171$, $\Delta t = 8.0959 \times 10^5$ years.
1109: Red solid line: $z = 24.148101$, $\Delta t = 1.9034 \times 10^5$ years.
1110: Black short dashed-long dashed line: $z = 24.132556$, $\Delta t = 1.2528 \times 10^5$ years.
1111: Black dot long-dashed line: $z = 24.126839$, $\Delta t = 4.6131 \times 10^4$ years.
1112: Black dot short-dashed line: $z = 24.124178$, $\Delta t = 2.1483 \times 10^4$ years.
1113: Black long-dashed line: $z = 24.121760$, $\Delta t = 1.9517 \times 10^4$ years.
1114: Black short-dashed line: $z = 24.120224$, $\Delta t = 1.2404 \times 10^4$ years.
1115: Black solid line: $z = 24.119543$, $\Delta t = 5.5024 \times 10^3$ years.
1116: }
1117: \label{fig.radprof_jlw0.1}
1118: \end{figure}
1119: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1120:
1121:
1122:
1123:
1124: %%%%%%%%%%%%%%%%%%% FIGURE %%%%%%%%%%%%%%
1125: \begin{figure}
1126: \begin{center}
1127: \includegraphics[width=0.3\textwidth]{f14a.eps}
1128: \includegraphics[width=0.3\textwidth]{f14b.eps}
1129: \includegraphics[width=0.3\textwidth]{f14c.eps}
1130: \includegraphics[width=0.3\textwidth]{f14d.eps}
1131: \includegraphics[width=0.3\textwidth]{f14e.eps}
1132: \includegraphics[width=0.3\textwidth]{f14f.eps}
1133: \end{center}
1134: \caption{
1135: Evolution of several spherically-averaged baryon quantities as a function
1136: of time for the J$_{21} = 1$ simulation.
1137: Panel (a): Number density as a function of radius.
1138: Panel (b): Enclosed gas mass as a function of radius.
1139: Panel (c): Temperature as a function of radius.
1140: Panel (d): $H_2$ fraction as a function of radius.
1141: Panel (e): Ratio of cooling time to sound crossing time as a function of radius.
1142: Panel (f): Ratio of cooling time to dynamical time as a function of radius.
1143: All quantities except enclosed gas mass are mass-weighted.
1144: Line types and weights correspond to different times, as follows.
1145: Red short dashed-long dashed line: $z = 18.384605$, $t = 1.995 \times 10^8$ years.
1146: Red dot long-dashed line: $z = 17.737221 $, $\Delta t = 5.2141 \times 10^6$ years.
1147: Red dot short-dashed line: $z = 17.394373$, $\Delta t = 1.7253 \times 10^6$ years.
1148: Red long-dashed line: $z = 17.373929$, $\Delta t = 3.6030 \times 10^5$ years.
1149: Red short-dashed line: $z = 17.330286$, $\Delta t = 7.7255 \times 10^5$ years.
1150: Red solid line: $z = 17.325077$, $\Delta t = 9.25088 \times 10^4$ years.
1151: Black short dashed-long dashed line: $z = 17.323130$, $\Delta t = 3.4610 \times 10^4$ years.
1152: Black dot long-dashed line: $z = 17.322398$, $\Delta t = 1.3001 \times 10^4$ years.
1153: Black dot short-dashed line: $z = 17.322050$, $\Delta t = 6.1828 \times 10^3$ years.
1154: Black long-dashed line: $z = 17.321880$, $\Delta t = 3.0262 \times 10^3$ years.
1155: Black short-dashed line: $z = 17.321796 $, $\Delta t = 1.4813 \times 10^3$ years.
1156: Black solid line: $z = 17.321757 $, $\Delta t = 7.0496 \times 10^2$ years.
1157: }
1158: \label{fig.radprof_jlw1em21.1}
1159: \end{figure}
1160: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1161: \clearpage
1162:
1163: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1164: \section{Neglected Physics and Possible Numerical issues}\label{sect.issues}
1165:
1166: In this paper we have examined aspects of the formation of Population III
1167: stars in the presence
1168: of a soft ultraviolet background, using an adaptive mesh refinement cosmological
1169: structure formation code. Given the nature of the tool and current limitations in
1170: computing power, some physics were neglected, and possible numerical issues may
1171: arise. We discuss this here.
1172:
1173: The primordial chemistry model used in these calculations ignores the effects of
1174: deuterium, lithium, and the various molecules that form between these elements
1175: and ordinary hydrogen. Deuterium and lithium have been shown to be unimportant
1176: in the temperature and density regimes that we have examined in this paper
1177: \markcite{1998A&A...335..403G,2002P&SS...50.1197G,2005MNRAS.361..850L}({Galli} \& {Palla} 1998, 2002; {Lipovka}, {N{\'u}{\~n}ez-L{\'o}pez}, \& {Avila-Reese} 2005).
1178: However, it is possible that they may be relevant
1179: in other situations of importance to Population III star formation -- in particular,
1180: regions which have been ionized to very high electron fractions may experience
1181: significant cooling from the HD molecule, which due to its permanent dipole moment
1182: makes it more than 100 times more effective as a cooling agent than molecular
1183: hydrogen (per molecule), and has the potential to cool gas down to approximately
1184: the temperature of the cosmic microwave background, which scales with redshift
1185: as $T_{cmb}(z) = 2.73 \times (1+z)$~K
1186: ~\markcite{2000MNRAS.314..753F,2002P&SS...50.1197G,2005MNRAS.361..850L}({Flower} {et~al.} 2000; {Galli} \& {Palla} 2002; {Lipovka} {et~al.} 2005).
1187: This gives a minimum baryon temperature of approximately $55$ Kelvin at $z=20$ and
1188: could reduce the minimum accretion rate onto a primordial protostar significantly.
1189: Lithium, while in principle an effective coolant as LiH, is safely ignored, since only
1190: a tiny fraction of lithium is converted into LiH~\markcite{2005PASJ...57..951M}({Mizusawa}, {Omukai}, \& {Nishi} 2005).
1191: A final chemical process that is omitted from our calculation is heating
1192: caused by molecular hydrogen formation at high densities (n $\ga 10^8$~cm$^{-3}$),
1193: the inclusion of which may result in differences in the temperature evolution
1194: of the gas in the highest density regimes considered here. This does not
1195: significantly affect the conclusions of our paper, as the significant differences
1196: we observe between simulations occur at densities much lower than
1197: $\sim 10^8$~cm$^{-3}$.
1198:
1199: Self-shielding of the photodissociating background by molecular hydrogen in the high-density
1200: gas is ignored in these calculations. Though this effect could
1201: in principle be important, the actual column
1202: densities of molecular hydrogen are typically far too small to actually block the soft UV flux.
1203: According to
1204: \markcite{2001ApJ...548..509M}{Machacek} {et~al.} (2001) (and references therein), a column density of
1205: $5 \times 10^{14}$~cm$^{-2}$ is enough for
1206: shielding to become important. However,
1207: this was derived for a static distribution of H$_2$, while the LW band consists of
1208: hundreds of individual lines
1209: whose width in this case is dominated by Doppler broadening. It is useful to note in
1210: this case that the average line
1211: width is $\sim 2$~km/s and the RMS baryon velocity in our calculations are $\sim 4$~km/s.
1212: In order for self-shielding to be important in the case of a turbulent medium, the column
1213: density must be much higher.
1214: Typical maximum H$_2$ column densities in our calculations are on the order of
1215: $10^{16}$~cm$^{-2}$, but these
1216: occur late in the collapse of the halo core, and in the highest density regions the cooling and
1217: H$_2$ production times are much shorter than the photodissociation time scale, at which point
1218: self-shielding becomes unimportant. It is worth noting that there are some regimes where
1219: self-shileding can be critical. \markcite{2007ApJ...659..908S}{Susa} (2007) finds that
1220: self-shielding can strongly affect the evolution of a collapsing primordial halo which is being
1221: illuminated with photodissociating flux by a neighboring star, though the situation is somewhat
1222: idealized.
1223: Additionally, we do not consider the more complicated effects relating to Population III
1224: stars which form in halos in the cosmic neighborhoods where previous generations of stars have
1225: existed. This allows us to ignore complex radiative, chemical and dynamical effects that would
1226: vastly complicate our
1227: calculations~\markcite{2005ApJ...628L...5O,2006astro.ph.12254J,2006astro.ph.10819Y,2007arXiv0705.3048G,
1228: 2007ApJ...659L..87A,2007MNRAS.375..881A}({O'Shea} {et~al.} 2005a; {Johnson}, {Greif}, \& {Bromm} 2006; {Yoshida} {et~al.} 2006a; {Greif} {et~al.} 2007; {Abel}, {Wise}, \& {Bryan} 2007; {Ahn} \& {Shapiro} 2007).
1229:
1230: A further effect that is ignored in this paper is $H^-$ photodetachment. This could in principle
1231: be a significant effect -- $H^-$ is the catalyst for molecular hydrogen formation in the
1232: dominant \h2 formation channel at densities $\la 10^8$~cm$^{-3}$, and it can be detached by
1233: photons with energies $h \nu \ga 0.75$~eV. Photons with this energy would be produced in
1234: great numbers by the same massive Population III stars that we assume are producing the
1235: molecular hydrogen photodissociating background. However, as shown by \markcite{2001ApJ...548..509M}{Machacek} {et~al.} (2001),
1236: the rate of photodetachment is orders of magnitude smaller than the rate of $H^-$ formation
1237: at densities comparable to that found in the centers of the halos examined in this work,
1238: and thus the process of $H^-$ photodetachment can be safely ignored.
1239:
1240: In this paper, we examine the effects of molecular hydrogen dissociating backgrounds which range in
1241: strength from J$_{LW} = 10^{-24}$ to $10^{-21}$~ergs$^{-1}$~cm$^{-2}$~Hz$^{-1}$~sr$^{-1}$ in the LW
1242: band ($11.18 - 13.6$ eV). This range is in good agreement with the expected range of photodissociating
1243: backgrounds predicted by~\markcite{2005ApJ...629..615W}{Wise} \& {Abel} (2005), but it is still finite. Examination of our
1244: simulations show that the case with the lowest UV background (J$_{21} = 10^{-3}$) is almost identical
1245: to the ``control'' (J$_{21} = 0$) calculation, justifying our choice of minimum value. Our upper value
1246: is consistent with Wise \& Abel, and examination of Figures~\ref{fig.halovals1} and
1247: \ref{fig.radprof_zcoll.1}--\ref{fig.radprof_zcoll.3}
1248: suggests that increases in the strength of the UV background (within reasonable values) will
1249: result in a further delay in halo
1250: collapse, but no major change in the mode of star formation observed. This is an effect of the halo
1251: properties -- a small amount of molecular hydrogen will always exist in these halos, allowing
1252: the gas at the center to cool and contract quasi-statically to higher densities.
1253: Once the gas has collapsed to very high densities, rapid H$_2$ formation via the 3-body process will
1254: occur, essentially independent of the strength of the soft UV background, and the
1255: gas will be able to cool down to $\simeq 200$ K very quickly.
1256:
1257: The effects of magnetic fields are completely ignored in the simulations discussed in this work.
1258: Magnetic fields are
1259: discussed in detail in Paper I, but, to summarize, a fairly high seed magnetic field is needed to
1260: be dynamically
1261: significant at the relatively low densities we explore in this work. The possible importance of magnetic
1262: fields has been explored in analytic and semi-analytic work
1263: \markcite{TanMcKee2004,2006MNRAS.371..444S,2007arXiv0704.1853M}({Tan} \& {McKee} 2004; {Silk} \& {Langer} 2006; {Maki} \& {Susa} 2007).
1264: We will examine the possible evolution of magnetic fields within the context of
1265: cosmological AMR simulations in a
1266: later paper (O'Shea \& Turk 2007, in preparation).
1267:
1268: The simulations presented in this paper are generated assuming a cosmology
1269: that is somewhat different than the currently-favored WMAP Year III ``best-fit''
1270: model~\markcite{2006astro.ph..3449S}({Spergel} {et~al.} 2006). Most importantly, our value of $\sigma_8$
1271: is 0.9, which is significantly higher than the WMAP value of 0.761. The general
1272: effect of a higher $\sigma_8$ is to cause structure formation to take place earlier,
1273: and thus at a given redshift one would expect significantly more halos in our chosen
1274: cosmological model than in the WMAP Year III model. However, the evolution of any
1275: single cosmological halo, such as the one examined in this work, is not particularly
1276: affected by this parameter in the sense that we are not examining halo stastical
1277: properties. In addition to $\sigma_8$, the ratio of $\Omega_b/\Omega_m$ in our
1278: simulations is 0.1337, while it is $0.1746$ in the WMAP Year III cosmology.
1279: This may result in some small quantitative differences in, e.g., the redshift of
1280: halo collapse, but should not significantly affect our results.
1281:
1282: We direct the reader to Paper I for a detailed discussion of other possible numerical issues,
1283: such as the generation of cosmological initial conditions, the assumption that the cold dark
1284: matter model is correct, the choice of halo in our simulations, and the size of the simulation
1285: volumes used.
1286:
1287: \section{Discussion}\label{discuss}
1288:
1289: This paper explores the formation of Population III stars in simulations with a constant
1290: soft UV background. Our results agree well qualitatively with that of \markcite{2001ApJ...548..509M}{Machacek} {et~al.} (2001);
1291: we both find that a soft UV background can delay the formation of molecular hydogen, and thus the
1292: cooling and collapse, of small ($\sim 10^6$~M$_\odot$) cosmological halos in which Population III
1293: stars form. We also find that increasing the soft UV background increases the minimum halo mass
1294: required for a halo to collapse (in a way similar to that of smoothing the dark matter
1295: power spectrum at small scales -- see \markcite{oshea_wdm}{O'Shea} \& {Norman} (2006) for a discussion). Machacek et al. derived
1296: a mass threshold for collapse as a function of the LW background flux that agrees well with our
1297: simulations, though the halo masses in our calculation are significantly higher. This is due to the halo that
1298: we examine being a ``typical'' halo rather than at the threshold mass for star formation. Presumably, if
1299: we performed these calculations using many halos in a range of cosmological realizations, we would find
1300: a minimum halo mass that is somewhat lower than the masses seen in our calculations.
1301: Our work is a significant improvement upon that of Machacek et al. in some respects, as our
1302: simulations are much more
1303: highly resolved and we examine the evolution of a single halo over a much wider range of
1304: soft UV background fluxes. One drawback of our work compared to Machacek et al. is that we examine the
1305: evolution of a single halo, albeit with a broad range of UV backgrounds.
1306:
1307: Our work is similar to the calculations in~\markcite{2003ApJ...592..645Y}{Yoshida} {et~al.} (2003) which examine
1308: Population III star formation in the presence of a photodissociating ultraviolet background.
1309: They find
1310: that for values of J$_{LW}$ above $10^{-23}$~erg s$^{-1}$~cm$^{-2}$~Hz$^{-1}$~sr$^{-1}$, hydrogen
1311: molecules are rapidly dissociated and gas cooling is inefficient, implying that
1312: the collapse of gas in the halo center may be delayed until atomic line cooling dominates, in halos with
1313: T$_{vir} \ga 10^4$~K. While we see delays in halo collapse, we do see that a tiny
1314: amount of \h2 can form even at high FUV background strengths, eventually
1315: allowing gas to cool and collapse to high densities without ever reaching temperatures
1316: at which atomic line cooling would be effective (T $\ga 10^4$~K).
1317: Yoshida et al. make this statement based
1318: on a set of simulations using two different values of J$_{LW}$ ($10^{-23}$ and $10^{-22}$), which
1319: bracket the ``break'' in properties seen in our results. The simulations used to obtain this result
1320: had SPH particles with m$_{gas} = 100.0$~h$^{-1}$~M$_\odot$ and thus an effective mass
1321: resolution comparable
1322: to the mass of gas within the quasistatically-collapsing central region of a halo. Our mass resolution is more than
1323: two orders of magnitude higher in the center of the cosmological halos that we examine.
1324: The differences between our result and theirs is likely due to the significant differences in
1325: resolution and may in fact be consistent once resolution is taken into account.
1326:
1327: We see a strong relationship between accretion rates onto the protostellar cloud
1328: and the strength of the photodissociating background. Calculations with higher
1329: ultraviolet background strengths typically have a larger spherically-averaged baryon accretion rate
1330: onto the evolving protostellar cloud. This is clearly due to variation in halo central
1331: temperature relating to the amount of molecular hydrogen existing in the center of the halo at the
1332: epoch of collapse. Accretion onto the evolving protostellar cloud is subsonic, and
1333: thus regulated by the local sound speed (which scales as T$^{0.5}$).
1334: This implies some relationship
1335: between the strength of the photodissociating background and the final stellar mass of
1336: the primordial star -- however, the details of this relationship are unclear, and depends
1337: on many factors. For example,~\markcite{2003ApJ...589..677O}{Omukai} \& {Palla} (2003) suggest that an increase in accretion
1338: rate above $\dot{m} \simeq 4 \times 10^{-3}$~M$_\odot$/yr
1339: will actually result in a \emph{decrease} in the final mass of the star due to radiative
1340: feedback from the evolving protostar. However, their results
1341: use one-dimenstional, fairly idealized models, and geometrical effects may be important. This is
1342: explored in~\markcite{TanMcKee2004}{Tan} \& {McKee} (2004), who use a combination of analytic and semi-analytic models of
1343: the evolving system. They suggest that accretion onto the protostar is highly non-spherical,
1344: and is in fact mediated by an accretion disk. They argue that a larger accretion rate onto the disk
1345: will lead to a larger star, though with some limitations. Finally, \markcite{2001ApJ...546..635O}{Omukai} (2001)
1346: and \markcite{2003ApJ...599..746O}{Omukai} \& {Yoshii} (2003) study the IMF of
1347: stars that form in T$_{vir} \ga 10^4$~K metal-free protogalaxies in the presence of an H$_2$ photodissociating
1348: background. They conclude that photodissociation actually decreases the Jeans mass of the gas at high
1349: densities, and thus reduces the fragmentation mass scale of the clouds and presumably the stellar mass.
1350: They use one-dimensional, spherically symmetric simulations, however. We can explore this issue in
1351: more detail in later calculations, when appropriate models for shielding, cooling via $H^-$, and
1352: more advanced chemistry have been implemented into Enzo. This will be examined in a later paper.
1353: It is worth noting that the range of accretion rates observed in this situation are within
1354: the range of rates seen in the $\Lambda$CDM simulations discussed in Paper I. This implies that the
1355: mass ranges of the resulting stars will not be significantly different than in the
1356: ``standard'' Population
1357: III star formation scenario.
1358:
1359:
1360: We see no fragmentation of the high-density baryon core, up
1361: to a baryon number density of $n \simeq 10^{10}$~cm$^{-3}$, for the entire range of simulations
1362: explored in this paper. This is due to a combination of effects, but predominantly the poor
1363: cooling properties of molecular hydrogen. The gas is relatively hot ($\sim 1000$ Kelvin)
1364: and thus has a high sound speed, which helps to damp out perturbations in the halo center which would
1365: otherwise result in multiple fragments. This effect is exacerbated in simulations with
1366: high ultraviolet background strengths, as the temperatures (and thus the sound speeds)
1367: are generally higher.
1368: This result implies that there is no fundamental change in the mode of Population III
1369: star formation as halos grow, and that more massive ``proto-galactic'' halos, with T$_{vir}
1370: \ga 10^4$~K, will continue to form a single massive star per halo. These halos will have
1371: a much larger binding energy than the smaller, M $\sim 10^5-10^6$~M$_\odot$ halos which have
1372: traditionally been examined by numerical simulations, and implies that multiple generations of star formation
1373: may be able to take place in a single halo. This leads to the possibility of ``self-enrichment,'' where
1374: a single Population III star enriches the high-density center of the halo to metallicities high
1375: enough to change the cooling properties of the gas. This could cause a strong change in the IMF
1376: \markcite{2006ApJ...643...26S,2007ApJ...661L...5S}({Santoro} \& {Shull} 2006; {Smith} \& {Sigurdsson} 2007), and these objects could make a significant contribution
1377: to the reionization history of the universe~\markcite{2003ApJ...586....1M,2006MNRAS.373..128G}({Mackey}, {Bromm}, \& {Hernquist} 2003; {Greif} \& {Bromm} 2006). We will examine whether the high-density gas in the centers
1378: of these more massive halos fragments in a later work.
1379:
1380: The values of accretion rates onto the evolving protostellar cloud which are
1381: observed with increasing photodissociating
1382: background flux (which are always comparable to those observed in Paper I), in addition to the
1383: apparent lack of fragmentation of gas in the halo center up to
1384: densities of n$_h \sim 10^{10}$~cm$^{-3}$, implies that there is no major
1385: change in the mass range of Population III stars as halo masses increase to T$_{vir} \sim 10^4$~K.
1386: This is in mild disagreement with previous semianalytic results by \markcite{2002ApJ...569..558O}{Oh} \& {Haiman} (2002) and
1387: \markcite{2003ApJ...586....1M}{Mackey} {et~al.} (2003). This is apparently
1388: because previous authors assumed that \h2 cooling in ``second generation'' (T$_{vir} \ga 10^4$~K) objects
1389: is unimportant, while we observe that a small \h2 fraction, which forms in equilibrium with the UV
1390: background, can cool gas effectively to temperatures of $\simeq 1000$~K. We do
1391: agree with these authors that the majority of primordial star formation will take place in objects with
1392: T$_{vir} \sim 10^4$~K, the so-called ``Second generation'' objects~\markcite{2006MNRAS.368.1301M}({MacIntyre}, {Santoro}, \& {Thomas} 2006)
1393:
1394: Our work appears to contradict the results of \markcite{2006MNRAS.373..128G}{Greif} \& {Bromm} (2006), who use analytic and semianalytic
1395: techniques to follow the collapse of gas within halos of T$_{vir} > 10^4$. They find that,
1396: in their model, gas cools
1397: to $\sim 8000$~K via atomic line cooling, and then contracts nearly isothermally to high densities,
1398: allowing a molecular hydrogen fraction of f$_{H2} \sim 0.001$ to build up independent of local
1399: density and temperature. The gas in the center of the halo then cools and allows the gas to fragment
1400: on scales of $\sim 20$~M$_\odot$. We do not see this mode of star formation, possibly because
1401: we never generate the significantly ionized halos with T$ > 10^4$~K upon which their scenario depends.
1402: Rather, we simulate halos that in the cases of J$_{21} \geq 0.1$ are built up to T$_{vir} \sim 10^4$~K
1403: via mergers, and the gas is never ionized to a significant degree. One possible reason for
1404: the observed differences is that we ignore the formation of deuterium hydride and its effects on the
1405: cooling properties of primordial gas. It is possible that, in some contexts, the inclusion of HD may
1406: result in enhanced fragmentation. This will be examined in future studies. The scenario described by Oh \& Haiman may still in fact be able
1407: to occur, and deserves further detailed numerical study to determine the fate of the collapsing gas and
1408: the implications of the (presumably) extremely high accretion rates onto the forming protostellar clouds.
1409:
1410: Our results are similar to those shown by \markcite{2007arXiv0707.2059W}{Wise} \& {Abel} (2007),
1411: who perform similar simulations examining the collapse of gas in cosmological halos in the presence
1412: of a photodissociating background. They also find that, regardless of the strength of the UV background,
1413: the collapse of gas in the centers of the cosmological halos in question still occurs via
1414: H$_2$ cooling. This holds even for quite extreme examples, such as when the original electron
1415: fraction is set to unphysically low levels (which should suppress H$_2$ formation, at
1416: least temporarily).
1417:
1418: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1419: \section{Summary}\label{summary}
1420:
1421: In this paper we have performed a suite of high dynamical range (L$/\Delta x \sim 5 \times 10^8$) 3D adaptive
1422: mesh cosmological simulations of the formation of Population III
1423: stars in a $\Lambda$CDM universe in the presence of a molecular
1424: hydrogen photodissocating (``Lyman-Werner'') ultraviolet
1425: background.
1426: The purpose of these calculations is to understand the effect that the soft
1427: ultraviolet background has on the evolution of the gas in collapsing cosmological halos
1428: and to determine possible effects on the forming protostellar cloud. These simulations
1429: are all of a single cosmological realization, but with varied ultraviolet background strengths.
1430: Our principal results are as follows:
1431:
1432: 1. Our calculations show that, as the flux of ultraviolet radiation in the Lyman-Werner
1433: band is increased, Population III star formation in a given cosmological halo
1434: is delayed to later times and, as a result, an increase in the virial mass of this
1435: halo at the onset of baryon collapse. This is in good
1436: agreement with previous work by~\markcite{2001ApJ...548..509M}{Machacek} {et~al.} (2001) and~\markcite{2003ApJ...592..645Y}{Yoshida} {et~al.} (2003).
1437:
1438: 2. We find that, contrary to previous work, the formation of primordial stars is never completely suppressed,
1439: regardless
1440: of the strength of the UV background. A small amount of molecular
1441: hydrogen always exists in cosmological halos, and allows cooling and halo collapse
1442: in gas which is bathed in a strong photodissociating background but has not
1443: reached $10^4$~K. The previously-suggested mode of star formation in ``second generation'' halos, where
1444: the collapse of gas to high density is completely suppressed until cooling can take place via atomic
1445: hydrogen line transitions, is never observed in our calculations, which have
1446: photodissociating background strengths up to J$_{21} = 1$.
1447:
1448: 3. Though the molecular hydrogen fraction in the centers of halos which are bathed in strong
1449: (J$_{21} \geq 0.1$) Lyman-Werner radiation is very small at early times (f$_{H2} \sim 10^{-6}-10^{-5}$), the
1450: gas at the center of the halo can still collapse due to the efficient cooling of molecular hydrogen at
1451: $2-5 \times 10^3$~K, and to the extended time these halos require to collapse compared to
1452: halos in the presence of much smaller UV backgrounds.
1453:
1454: 4. We observe that the estimated accretion rate onto the forming protostellar
1455: cloud varies strongly as a function of J$_{LW}$, with simulations that have
1456: a stronger ultraviolet background having higher accretion rates. This is
1457: a function of the suppression of molecular hydrogen formation
1458: (and thus suppression of effective cooling) by the photodissociating background
1459: and of the higher virial temperatures of these halos when the gas finally collapses.
1460:
1461: 5. Only a single clump is formed at the center of each collapsing cosmological
1462: halo, regardless of the strength of the photodissociating background, up to
1463: a baryon number density of $n \sim 10^{10}$~cm$^{-3}$. This implies that, as in
1464: the more commonly-studied mode of star formation, we will find only a single star
1465: per halo even in objects which are massive enough that T$_{vir} \ga 10^4$~K.
1466:
1467:
1468: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1469: \acknowledgments{
1470: B.W.O. would like to thank Tom Abel, Greg Bryan, Simon Glover, Thomas Greif,
1471: Matthew Turk, and John Wise for
1472: useful discussions. This work supported in part by NASA
1473: grant NAG5-12140 and NSF grant AST-0307690.
1474: B.W.O. carried out this work under the auspices of the
1475: National Nuclear Security Administration of the
1476: U.S. Department of Energy at Los Alamos National
1477: Laboratory under Contract No. DE-AC52-06NA25396, and was
1478: supported by a LANL Director's Postdoctoral Fellowship (DOE LDRD grant
1479: 20051325PRD4).
1480: The simulations were performed at SDSC and NCSA with computing time provided by
1481: NRAC allocation MCA98N020.
1482: }
1483:
1484: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1485: %\bibliographystyle{apj}
1486: \bibliography{} % looks in ms.bib for bibliography info
1487:
1488: \end{document}
1489:
1490:
1491: