1: \documentclass[12pt,preprint]{aastex}
2: %\documentclass{emulateapj}
3: %\documentclass[aps,preprint]{revtex4}
4: %\usepackage{graphicx}
5: %\usepackage{amssymb}
6: %\usepackage{epstopdf}
7:
8:
9: \newcommand{\fft}[2]{\frac{#1}{#2}}
10: \newcommand{\ft}[2]{{\textstyle\frac{#1}{#2}}}
11:
12: \long\def\symbolfootnote[#1]#2{\begingroup%
13: \def\thefootnote{\fnsymbol{footnote}}\footnote[#1]{#2}\endgroup}
14:
15:
16: \DeclareGraphicsRule{.tif}{png}{.png}{`convert #1 `dirname #1`/`basename #1 .tif`.png}
17:
18: \begin{document}
19:
20: \title{Stellar Structure of Dark Stars: a first phase of Stellar Evolution
21: resulting from Dark Matter Annihilation}
22: \author{
23: Katherine Freese\altaffilmark{1},
24: Peter Bodenheimer\altaffilmark{2},
25: Douglas Spolyar\altaffilmark{3},
26: and
27: Paolo Gondolo\altaffilmark{4}}
28:
29: \email{ktfreese@umich.edu, peter@ucolick.org, dspolyar@physics.ucsc.edu, paolo@physics.utah.edu}
30:
31: \altaffiltext{1}{Michigan Center for Theoretical Physics, Physics Dept.,
32: Univ. of Michigan, Ann Arbor, MI 48109}
33: \altaffiltext{2}{Dept. of Astronomy and Astrophysics, University of California,
34: Santa Cruz, CA 95064}
35: \altaffiltext{3}{Physics Dept., University of California, Santa Cruz, CA 95064}
36: \altaffiltext{4}{Physics Dept., University of Utah, Salt Lake City, UT 84112}
37:
38: \begin{abstract}
39: %\noindent
40: Dark Stars are the very first phase of stellar evolution in the
41: history of the universe: the first stars to form (typically at redshifts $z \sim
42: 10-50$) are powered by heating from dark matter (DM) annihilation
43: instead of fusion (if the DM is made of particles which are their own
44: antiparticles). We find equilibrium polytropic configurations for
45: these stars; we start from the time DM heating becomes important ($M
46: \sim 1-10~M_\odot$) and build up the star via accretion up to 1000~
47: M$_\odot$. The dark stars, with an assumed particle mass of 100 GeV,
48: are found to have luminosities of a few times $10^6$ L$_\odot$,
49: surface temperatures of 4000--10,000 K, radii $\sim 10^{14}$ cm,
50: lifetimes of at least $ 0.5$ Myr, and are predicted to show lines of
51: atomic and molecular hydrogen. Dark stars look quite different from
52: standard metal-free stars without DM heating: they are far more massive
53: (e.g. $\sim 800 M_\odot$ for 100 GeV WIMPs), cooler, and larger,
54: and can be distinguished in future observations, possibly even by JWST or TMT.
55:
56: \end{abstract}
57: \keywords{Dark Matter}
58:
59: \section{Introduction}
60: The first stars in the Universe mark the end of the cosmic dark ages,
61: reionize the Universe, and provide the enriched gas required for later
62: stellar generations. They may also be important as precursors to
63: black holes that coalesce and power bright early quasars. The first
64: stars are thought to form inside dark matter (DM) halos of mass $ 10^5
65: M_\odot$--$ 10^6 M_\odot$ at redshifts $z \sim 10-50$ (Yoshida et
66: al. 2003). These halos consist of 85\% DM and 15\% baryons in the
67: form of metal-free gas made of H and He. Theoretical calculations
68: indicate that the baryonic matter cools and collapses via H$_2$
69: cooling (Peebles \& Dicke 1968, Matsuda et al. 1971, Hollenbach \&
70: McKee 1979) into a single small protostar (Omukai \& Nishi 1998) at
71: the center of the halo (for reviews see Ripamonti \& Abel 2005;
72: Barkana \& Loeb 2001; Bromm \& Larson 2004).
73:
74: Previously, Spolyar et al. (2008; hereafter, Paper I) first considered the
75: effect of DM particles on the first stars during their
76: formation. Any DM particle which is capable of annihilating with
77: itself in such a way as to give the correct relic abundance today will
78: also annihilate wherever the DM density is high.
79: The first protostars and stars are particularly good sites for annihilation
80: because they form at high redshifts (density scales as $(1+z)^3$) and
81: in the high density centers of DM haloes. Paper I found that
82: DM annihilation provides a powerful heat source in the first stars, a
83: source so intense that its heating overwhelms all cooling mechanisms.
84: Paper I suggested that the
85: very first stellar objects might be {\it Dark Stars} (DS), a new phase of
86: stellar evolution in which the DM -- while only a negligible fraction
87: of the star's mass -- provides the key power source for the star through
88: DM heating. Note that the term 'Dark' refers to the power source, not the
89: luminosity. In this paper, we continue the work originally suggested
90: in Paper I by studying the DS structure.
91: %We find their mass, luminosity, temperature, and radius.
92:
93: The canonical example of particle DM is Weakly Interacting Massive
94: Particles (WIMPs), which automatically provide the right amount of DM,
95: i.e.\ $\sim$ 24\% of the current energy density of the Universe. In
96: many theories WIMPs are their own antiparticles and annihilate with
97: themselves in the early universe, leaving behind this relic
98: density. In particular, the neutralino, the supersymmetric partner of
99: the W, Z, and Higgs bosons, is a strong candidate (reviewed by Jungman
100: et al. 1996). As our canonical values, we use the standard
101: $\langle \sigma v \rangle = 3 \times 10^{-26}\,{\rm cm^3/s}$ for the
102: annihilation cross section and $m_\chi = 100\,{\rm GeV}$ for the
103: particle mass. A companion paper will generalize to other masses and
104: cross sections. The analysis in this paper could apply equally well to
105: other DM candidates.
106:
107: WIMP annihilation produces energy at a rate per unit volume
108: \begin{equation}
109: \hat Q_{DM} = \langle \sigma v \rangle \rho_\chi^2/m_\chi ,
110: \label{eq:Q}
111: \end{equation}
112: where $\rho_\chi$ is the energy density of the WIMPs.
113: In the early stages of Pop III star formation, when the gas density is
114: low ($n \lesssim 10^4 {\rm cm}^{-3}$), most of the annihilation products simply
115: escape from the protostar without heating it
116: (Ripamonti et al. 2007). However,
117: a crucial transition takes place (Paper I) when the gas density
118: of the collapsing protostar exceeds a critical value at which point
119: most of the annihilation energy is trapped in the star. For a 100
120: GeV particle, at hydrogen density $\sim 10^{13}\, {\rm
121: cm}^{-3}$,
122: typically 1/3 of the energy is lost to neutrinos that escape the star,
123: while the other 2/3 of the energy is trapped inside the star.
124: Hence the luminosity from the DM heating is
125: \begin{equation}
126: \label{DMheating}
127: L_{DM} \sim {2 \over 3} \int \hat Q_{DM} dV
128: \end{equation}
129: where $dV$ is the volume element.
130:
131: The properties of the collapsing protostellar clouds have been given
132: by 3D simulations (Abel et al. 2002; Gao et al. 2007). At the time
133: when the density reaches $n = 10^{13}$cm$^{-3}$, the critical value
134: for 100 GeV particles, Paper I found a proto-DS in equilibrium with a
135: radius of 17 AU and a mass of $0.6 M_\odot$, giving a DM luminosity of
136: $\sim 140$ L$_\odot$. As more mass accretes onto the DS, the
137: protostellar luminosity begins to exceed the DM heating, so that the
138: protostar is no longer in thermal equilibrium. Thus it must contract
139: which increases the DM density until the DM heating as given in
140: equation (\ref{DMheating}) matches its radiated luminosity.
141: %(hydrodynamic evolution in preparation).
142:
143: In this calculation, we assume that such a situation can be reached,
144: and we then build up the Dark Star from a few solar masses up to 1000
145: $M_\odot$, finding its structure as a polytrope in hydrostatic and
146: thermal equilibrium at each step in mass. As we build up the star
147: more DM is pulled into the star via adiabatic contraction and
148: subsequently annihilates; we find that the annihilation fuel contained
149: in the star can thereby last $\sim 10^6$yr. While the results of this
150: paper were being written, a paper appeared by Iocco et al. (2008)
151: which included DM heating in Pop III pre-main-sequence evolution of a
152: set of stars of {\it fixed mass}, finding that the quasi-hydrostatic
153: contraction is halted for times of $2 \times 10^3$ ($2 \times 10^4$)
154: yr for stars of mass 600 (9) M$_\odot$, at radii $\approx$ a few AU.
155:
156: During the evolution of a DS, additional WIMPs could be captured via
157: scattering off of nuclei. The cross section for scattering
158: ($\sigma_s$) is very uncertain. For $\sigma_s <10^{-39}$ cm$^2$ we
159: find that a DM particle undergoes less than one scattering event in 1
160: Myr in the evolutionary stage considered in this paper. The
161: experimental bounds for 100 GeV particles from DM searches are
162: $\sigma_s \lesssim 2 \times 10^{-43}$ cm$^2$ for the spin-independent
163: case (Gaitskill et al. 2008) and $\sigma_s \lesssim 3.5 \times
164: 10^{-39}$ cm$^2$ for the spin-dependent case (Savage et al. 2004).
165: Hence we assume negligible scattering here. However at later stages of
166: the evolution, once the DM density becomes too low to support the star
167: via heating, the DS contracts until nuclear burning sets in. At these
168: higher densities scattering at the experimentally allowed limit would
169: become important. DM passing through the star could be captured and
170: again drive DM heating. These effects have been considered for
171: main-sequence and pre-main-sequence DS (Freese et al. 2008; Iocco
172: 2008; Iocco et al. 2008), who find that the DM heating could dominate
173: nuclear fusion as long as the background DM density (from which the capture
174: takes place) remains high enough. Future work will further consider scattering in the
175: DS.
176:
177: We also cite previous work on DM annihilation in today's stars
178: (less powerful than in the first stars): Krauss et al (1985);
179: Bouquet \& Salati (1989); Salati \& Silk (1989); Moskalenko \& Wai (2007);
180: Scott et al. (2007); Bertone \& Fairbairn (2007).
181:
182: \section{Equilibrium Structure}
183:
184: We make the assumption that the dark
185: stars (DS) can be described as polytropes in hydrostatic equilibrium
186: \begin{equation}
187: \label{eq:polytrope}
188: P = K \rho^{1 + 1/n} .
189: \end{equation}
190: where $P$ is the pressure, $\rho$ is the density, and the constant $K$ is
191: determined once the total mass and radius are specified (Chandrasekhar
192: 1939). Pre-main-sequence stellar models are adequately described by polytropes
193: in the range $n=1.5$ (fully convective) to $n=3$ (fully radiative).
194: For a given stellar mass, we iterate the radius of the model
195: to find the point of thermal equilibrium, that is,
196: the total DM heating matches the radiated luminosity.
197: We then add 1 M$_\odot$, calculate a new equilibrium, and continue up to
198: 1000 M$_\odot$. In the standard scenario
199: of formation of the first stars, it was found that at $n \sim 10^4$
200: cm$^{-3}$, the mass of the protostellar cloud exceeds the Jeans
201: mass (e. g. Bromm, Coppi, \& Larson 2002). This amount of baryonic material,
202: $\sim 1000 $ M$_\odot$, could fall down
203: onto the DS and in the process bring in more DM with it.
204: %In fact, even more matter could rain down, since the
205: %initial halo contains $\sim 10^5 M_\odot$
206: %in baryonic matter and $\sim 10^6 M_\odot$ in DM, but
207: %we stop the sequence at 1000 M$_\odot$.
208:
209:
210: \subsection{DM Densities}
211: The DM densities in the protostar are derived as described in Paper I.
212: We take a $10^6$ M$_\odot$ halo composed of 85\% DM and 15\% baryons.
213: We take an initial Navarro, Frenk, \& White profile (1996; NFW) with a
214: concentration parameter $c=2$ at $z=20$ in a standard $\Lambda$CDM
215: universe. We follow the DM response to the changing baryonic
216: gravitational potential as the protostellar gas condenses. As the
217: baryons come to dominate the potential well in the core, they pull the
218: DM particles inward. We use the simple adiabatic contraction method of
219: Blumenthal et al. (1986), Barnes \& White (1984), and Ryden \& Gunn
220: (1987) (hereafter Blumenthal method) to estimate the resultant DM
221: density profile. The method has the limitation that all halo
222: particles are taken to be on circular orbits. Recently (Freese et
223: al. 2008b), we did an exact calculation using an algorithm originally
224: developed by Young (1980) which takes into account radial motions as
225: well. The results for the DM density agree with those from the
226: Blumenthal method to within a factor of 2. This factor of 2 may be
227: compensated by the fact that recent simulations by Via Lactea II
228: (Diemand et al. 2008) find initial DM density profiles that are
229: steeper in the inner core ($\rho_\chi \propto r^{-1.2}$ rather than
230: $\rho_\chi \propto 1/r$). Hence, the Blumenthal method should give
231: reasonable results. The DM density profile in the DS is calculated at
232: each iteration of the stellar structure, so that the DM luminosity can
233: be determined.
234:
235: \subsection{Basic Equations}
236:
237:
238: The basic equation is that of hydrostatic equilibrium
239: \begin{equation}
240: {dP \over dr} = - \rho {GM_r \over r^2}
241: \end{equation}
242: where ${dM_r \over dr} = 4 \pi r^2 \rho(r)$, $\rho(r)$ is the
243: total density (gas plus DM) at radius $r$, and $M_r$ is the enclosed mass
244: within radius $r$.
245: The temperature of the gas $(T(r)$) is determined from the equation of state
246: of a mixture of ideal gas and radiation:
247: \begin{equation}
248: \label{eq:eqnofstate}
249: P(r) = {{\rho k_B T(r)}\over {m_u \bar m}} + {1 \over 3} aT(r)^4
250: = P_g + P_{rad}
251: \end{equation}
252: where $k_B$ is Boltzmann's constant,
253: $m_u$ is the atomic mass unit, and the mean atomic weight $\bar m
254: = (2X + 3/4 Y)^{-1} =0.588$. We take the H mass fraction $X=0.76$
255: and the He mass fraction $Y=0.24$. In the resulting models
256: $T\gg10,\!000$ K except near the very surface,
257: so the H and He are ionized and the H$_2$ is dissociated.
258: We will find the radiation pressure to be important once the DS
259: becomes heavier than $\sim 100 M_\odot$.
260: We also require the DS to be in thermal equilibrium,
261: \begin{equation}
262: L_* = 4 \pi \sigma_B R_S^2 T_{\rm eff}^4 = L_{DM}
263: \label{eq:stellarlum}
264: \end{equation}
265: where $T_{\rm eff}$ is the effective surface temperature of the star at
266: its photospheric radius $R_S$. Note that the observable properties
267: are determined by this temperature.
268: The location of the photosphere may be determined roughly by the
269: requirement that the optical depth outside of $R_S$ is $\tau \sim 1$,
270: which is equivalent to using the photospheric boundary condition
271: $\kappa P= {2 \over 3} g$ where $g$ is the surface gravity. We use a
272: zero-metallicity Rosseland mean opacity ($\kappa$) table from OPAL
273: (Iglesias \& Rogers 1996), supplemented at the lowest temperatures by
274: opacities from Lenzuni et al. (1991).
275:
276: For each DS mass, we find the equilibrium star in the following way.
277: We guess a value for the outer radius $R_S$
278: which, along with the mass and the polytropic assumption,
279: can then be used to determine the baryon density distribution. Then
280: the Blumenthal method determines the DM density. One can then
281: use equations (\ref{eq:Q}) and (\ref{DMheating}) to find the amount of heating in each shell.
282: Our stellar code integrates outwards from the center of the DS, takes
283: a few hundred radial steps, and stops once it satisfies the photospheric
284: boundary condition. The temperature there is set to $T_{\rm eff}$.
285: Now, one compares $L_\ast$ with $L_{\rm DM}$.
286: If $L_{DM}<L_*$, then the next guess for $R_S$ must be smaller in order to
287: increa
288: boundary condition. The temperature there is set to $T_{\rm eff}$.
289: Now, one compares $L_\ast$ with $L_{\rm DM}$.
290: If $L_{DM}<L_*$, then the next guess for $R_S$ must be smaller in order to
291: increase $L_{DM}$ and at the same time decrease $L_\ast$.
292: Conversely, if $L_{DM} > L_\ast$, then the star must expand
293: in order to reduce the DM heating. We iterate to a convergence in the $L$'s
294: to 1 part in $10^4$.
295:
296: \subsection{Building up the Mass}
297:
298: Then we allow surrounding matter from the original baryonic core
299: to accrete onto the DS
300: at $2 \times 10^{-3}$
301: M$_\odot/{\rm yr}$, roughly $M_{\rm core}/t_{\rm ff}$, where $t_{\rm ff}$
302: is the free-fall time of the core.
303: The initial DS mass is 3 M$_\odot$ and the increment is
304: 1 M$_\odot$. We
305: remove the amount of DM that has annihilated at each stage at each
306: radius. We continue stepping up in mass until we
307: reach 1000 M$_\odot$,
308: the Jeans mass of the core (Bromm \& Larson 2004).
309:
310: With the above accretion rate, it takes $5 \times 10^5$ yr to build up
311: to 1000 M$_\odot$. Hence the lifetime of the DS is at least this
312: long. By this time a significant fraction of the DM inside the DS
313: has annihilated away. It is not
314: known whether or not the DM inside the DS can be repopulated from DM
315: particles in the $10^6 M_\odot$ halo surrounding it; this question
316: would require numerical resolution not currently available.
317:
318: In the future, it would be interesting to study the accretion process
319: in more detail. It is likely to proceed via the formation of a disk
320: with an accompanying accretion luminosity. In the standard Pop III
321: star formation process of accretion onto a small $10^{-3}M_\odot$
322: nugget, the luminosity has an accretion-driven phase; here, on the
323: other hand the accretion luminosity of the much larger DS is
324: always negligible. In any case our treatment of the structure of the
325: stellar interior is probably unchanged by the presence of the disk.
326: Previously McKee \& Tan (2007) have studied the role of angular
327: momentum in Pop III stars in the absence of DM. One should reconsider
328: angular momentum in the case of DS as well.
329:
330: \section{Results}
331:
332:
333:
334: Table 1 and Figure 1 illustrate our results for standard parameters
335: for $M_*=(10 - 1000) M_\odot$ and for n=1.5. In the Table we present the sequence
336: of central temperature $T_c$, photospheric
337: radius $R_S$, central gas density $\rho_c$, central DM density
338: $\rho_{\chi,c}$, stellar luminosity (equal to DM heating luminosity)
339: $L_*$, surface temperature $T_{\rm eff}$, total DM mass inside the star
340: $M_{DM}$, and time evolved since DM heating dominates inside the star.
341: At 1000 $M_\odot$ $\rho_c$
342: is far lower than for any metal-free Zero Age Main
343: Sequence star.
344: Figure 1 plots the baryon and DM density profiles.
345: The DM density is many of orders of
346: magnitude lower than the baryon density throughout the evolution
347: and yet the DM annihilation powers the star. As time goes on, one
348: can see that the DM is depleted in the interior regions of the star,
349: due to annihilation, and the density becomes very nearly constant.
350: The plot of $L_{\rm DM}(r)$, the dark matter luminosity integrated out to
351: radius $r$, shows that the heating is spread out over much of
352: the volume of the DS; thus it is not particularly sensitive to changes
353: in the details of the adiabatic contraction model.
354: By the time the DS reaches $1000$ M$_\odot$, the
355: amount of DM in the star is only 1/3 $M_\odot$, and 1/3 of the DM in
356: the DS has annihilated away. We also find that at each evolutionary state,
357: $L_\ast$ is typically
358: an order of magnitude less than the Eddington luminosity for a star
359: of that mass and radius.
360:
361: We have found that the DS is fully convective for stellar masses below
362: $100 M_\odot$; makes the transition from convective to radiative in
363: the $M_*=(100-200) M_\odot$ mass range (with a radiative zone growing
364: outward from the center); and then becomes (almost) fully radiative
365: (but for a small convective region at the surface) for $M_* > 200
366: M_\odot$. Hence the initial convective period is best described by an
367: $n$=3/2 polytrope while the later radiative period is best described
368: by an $n$=3 polytrope.
369:
370: The results above are for an $n$=3/2 polytrope all the way up to the
371: final mass. For an $n$=3 polytrope (more appropriate at the higher
372: masses), calculated up to 600 $M_\odot$, the results are
373: qualitatively the same. For $M_*=600 M_\odot$, the $n$=3 case gives
374: $T_{\rm eff}=9100K$, $R_S= 6.0 \times 10^{13}$cm, $L_*=4.6 \times 10^6
375: L_\odot$, and $T_c =2.2 \times 10^6K$; while the $n$=1.5 case gives
376: $T_{\rm eff}=6370K$, $R_S=1.0 \times 10^{14}$cm, $L_*=3.04 \times 10^6
377: L_\odot$, and $T_c = 6.88 \times 10^5K$. Thus the results for the
378: $n=1.5$ polytrope give the basic picture. Note that DS have much
379: lower $T_{\rm eff}$ than their standard metal free (Pop. III)
380: main-sequence counterparts in the absence of DM, which radiate at
381: $T_{\rm eff}>$30,000K. This difference gives a markedly different
382: observable signature for the DS than for the standard Pop III stars.
383:
384: \section{Conclusions}
385:
386: We have followed the growth of equilibrium Dark Stars,
387: powered by DM annihilation, up to 1000 M$_\odot$. The objects have
388: sizes of a few AU and central $T_c\approx 10^5-10^6$ K. Sufficient DM
389: is brought into the star by contraction from the DM halo to result in
390: a DS which lives at least 0.5 Myr (the lifetime could be significantly
391: longer if DM capture becomes important
392: at the later stages, as long as the background DM density is high
393: enough for capture to take place). Because of the relatively low $T_{\rm
394: eff}$ (4000--10,000 K), feedback mechanisms for shutting off accretion
395: of baryons, such as the formation of HII regions or the dissociation
396: of infalling H$_2$ by Lyman-Werner photons, are not effective. The
397: implication is that main-sequence stars of Pop. III are very massive.
398: This conclusion depends on uncertain parameters such as the DM
399: particle mass, the accretion rate, and scattering, effects that will
400: be studied in future work.
401:
402: Although DS shine with a few $10^6 L_\odot$ they would be very
403: difficult to observe at $z \sim 10-50$. One can speculate that
404: pristine regions containing only H and He might still exist to lower
405: redshifts; then DS forming in these regions might be easier to detect.
406: One may hope that the ones that form most recently are detectable by
407: JWST or TMT and differentiable from the standard metal-free Pop. III
408: objects. DS are also predicted to have atomic hydrogen lines
409: originating in the warmer photospheres, and H$_2$ lines arising from
410: the infalling material, which is still relatively cool.
411:
412: It has been argued that Pop III.1 stars (the very first metal-free
413: stars) may constitute at most $\sim 10\%$ of metal poor stars on
414: observational grounds. Heger \& Woosley (2002; HW) showed that for
415: $140 M_\odot < M < 260 M_\odot$, pair instability (SN) lead to
416: odd-even effects in the nuclei produced that are strongly constrained
417: by observations. Thus if Pop III.1 stars are really in this mass
418: range one would have to constrain their abundance. For $M > 260
419: M_\odot$, HW find that no SN occurs, and the end result of stellar
420: evolution is collapse of the entire star into a black hole. We
421: expect, based on extension of the $n=3$ calculation, that our DS runs
422: out of DM at about 700--900 M$_\odot$ (for $m_\chi = 100$ GeV). Then
423: it must contract to the main sequence, where nuclear burning sets in,
424: and further evolution would proceed as in HW. Alternatively, the
425: evolution could proceed as described by Ohkubo et al. (2006) who
426: found that metal-free stars of 500 and 1000 $M_\odot$, taking into
427: account two-dimensional effects, did blow up as SN, leaving about
428: half their mass behind in a black hole. In this case the SN might be
429: observable signatures of DS, distinguishable since they arise from
430: such high mass stars. The end product in either case would be a
431: plausible precursor of the otherwise unexplained $10^9 M_\odot$ black
432: holes at $z=6$ (Yoshida et al., in prep).
433:
434: We acknowledge support from: the DOE and MCTP via the Univ.\ of
435: Michigan (K.F.); NSF grant AST-0507117 and GAANN (D.S.); NSF grant
436: PHY-0456825 (P.G.). K.F. acknowledges the hospitality of the Physics
437: Dept. at the Univ. of Utah. K.F. and D.S. are extremely grateful to
438: Chris McKee and Pierre Salati for their encouragement of this line of
439: research, and to A. Aguirre, L. Bildsten, R. Bouwens, J. Gardner,
440: N. Murray, J. Primack, M. Rieke, C. Savage, J. Sellwood, J. Tan, and
441: N. Yoshida for helpful discussions.
442:
443: \begin{thebibliography}{99}
444:
445: \bibitem[]{567}
446: Abel, T., Bryan, G. L., \& Norman, M. L. 2002.
447: Science, 295, 93
448:
449: \bibitem[]{571}
450: Barkana, R, \& Loeb, A. 2001.
451: Phys.\ Rep., 349, 125
452:
453: \bibitem[]{575}
454: Barnes, J., \& White, S. D. M. 1984. MNRAS, 211, 753
455:
456: \bibitem[]{578}
457: Bertone, G., \& Fairbairn, M. 2007. arXiv:0711.1485
458:
459: \bibitem[]{581}
460: Blumenthal, G. R., Faber, S. M., Flores, R., \& Primack, J. R. 1986.
461: ApJ, 301, 27
462:
463: \bibitem[]{585}
464: Bouquet, A., \& Salati, P. 1989. ApJ, 346, 284
465:
466: \bibitem[]{586}
467: Bromm, V., Coppi, P. S., \& Larson, R. B. 2002,
468: ApJ, 564, 23
469:
470: \bibitem[]{588}
471: Bromm, V., \& Larson, R. B. 2004,
472: ARA\&A, 42, 79
473:
474: \bibitem[]{592} Chandrasekhar, S. 1939, An Introduction to the Study of
475: Stellar Structure (Chicago: Univ. of Chicago Press)
476:
477: \bibitem[]{595} Diemand, J., Kuhlen, M., Madau, P., Zemp, M., Moore, B., Potter, D.,
478: \& Stadel, J. 2008. arXiv:0805.1244
479:
480: \bibitem[]{598}
481: Freese, K., Spolyar, D., \& Aguirre, A. 2008a.
482: arXiv:0802.1724
483:
484: \bibitem[]{602}
485: Freese, K., Gondolo, P., Sellwood, J. A., \& Spolyar, D. 2008b.
486: arXiv:0805.3540
487:
488: \bibitem[]{606} Gao, L., Abel, T., Frenk, C. S., Jenkins, A.,
489: Springel, V., \& Yoshida, N. 2007. MNRAS, 378, 449
490:
491: \bibitem[]{609}
492: Gaitskill, R., Mandic, V., \& Filippini, J. 2008. {\tt http://dmtools.berkeley.edu/limitplots/}
493:
494: \bibitem[]{612}
495: Heger, A., \& Woosley, S. E. 2002.
496: ApJ, 567, 532 (HW)
497:
498: \bibitem[]{616}
499: Hollenbach, D., \& McKee, C. F. 1979.
500: ApJS, 41, 555
501:
502: \bibitem[]{620} Iglesias, C. A., \& Rogers, F. J. 1996. ApJ, 464, 943
503:
504: \bibitem[]{622}
505: Iocco, F. 2008. ApJ, 667, L1
506:
507: \bibitem[]{625}
508: Iocco, F., Bressan, A., Ripamonti, E., Schneider, R., Ferrara, A.,
509: \& Marigo, P. 2008.
510: arXiv:0805.4016
511:
512: \bibitem[]{630}
513: Jungman, G., Kamionkowski, M., \& Griest, K. 1996.
514: Phys.\ Rept., 267, 195
515:
516: \bibitem[]{634} Krauss, L., Freese, K., Press, W., \& Spergel, D. N. 1985.
517: Ap J, 299, 1001
518:
519: \bibitem[]{637} Lenzuni, P., Chernoff, D. F., \& Salpeter, E. 1991. ApJS,
520: 76, 759
521:
522: \bibitem[]{640}
523: Matsuda, T., Sato, H., \& Takeda, H. 1971.
524: Prog.\ Theor.\ Phys., 46, 416
525:
526: \bibitem[]{644}
527: McKee, C. F., \& Tan, J. C. 2007.
528: arXiv:0711.1377
529:
530: \bibitem[]{648}
531: Moskalenko, I. V., \& Wai, L. L. 2007.
532: ApJ., 659, L29
533:
534: \bibitem[]{652}
535: Navarro, J. F., Frenk, C. S., \& White, S. D. M. 1996.
536: ApJ, 462, 563
537:
538: \bibitem[]{654}
539: Ohkubo, T., Umeda, H., Maeda, K., et al. 2006.
540: ApJ, 645, 1352
541:
542: \bibitem[]{656}
543: Omukai, K., \& Nishi, R. 1998.
544: ApJ, 508, 141
545:
546: \bibitem[]{660}
547: Peebles, P. J. E., \& Dicke, R. H. 1968.
548: ApJ, 154, 891
549:
550: \bibitem[]{664}
551: Ripamonti, E., \& Abel, T. 2005.
552: arXiv:astro-ph/0507130.
553:
554: \bibitem[]{668}
555: Ripamonti, E., Mapelli, M., \& Ferrara, A. 2007.
556: MNRAS, 375, 1399
557:
558: \bibitem[]{672}
559: Ryden, B. S., \& Gunn, J. E. 1987. ApJ, 318, 15
560:
561: \bibitem[]{675}
562: Salati, P., \& Silk, J. 1989. ApJ, 338, 24
563:
564: \bibitem[]{678}
565: Savage, C., Gondolo, P., \& Freese, K. 2004. Phys. Rev. D,
566: 70, 123513
567:
568: \bibitem[]{682}
569: Scott, P., Edsjo, J., \& Fairbairn, M. 2007.
570: arXiv:0711.0991
571:
572: %\bibitem[]{686}
573: %Spergel, D. N., \& Press, W. H. 1985. ApJ, 294, 663
574:
575: \bibitem[]{689}
576: Spolyar, D., Freese, K., \& Gondolo, P. 2008.
577: Phys. Rev. Lett., 100, 051101
578:
579: \bibitem[]{693} Yoshida, N., Abel, T., Hernquist, L., \& Sugiyama, N. 2003.
580: ApJ, 592, 645
581:
582: \bibitem[]{696}
583: Young, P. 1980. ApJ, 242, 1232
584:
585:
586:
587: \end{thebibliography}
588:
589: \clearpage
590:
591: \begin{table*}
592: \caption{ Properties and Evolution of Dark Stars for $m_\chi = 100$
593: GeV,
594: $\dot M = 2 \times 10^{-3} M_\odot/{\rm yr}$, $\langle \sigma v \rangle = 3 \times 10^{-26}$ cm$^3$/s,
595: polytropic $n= 1.5$.
596: %\vspace{0.1 in}
597: }
598: \begin{center}
599: {
600: \small
601: \begin{tabular}{||l|l|l|l|l|l|l|l|l|c||}
602: \hline\hline
603: $M_*$ & $T_c$ & $R_S $ & $\rho_c $ &
604: $\rho_{\chi,c} $ & $L_* $ & $T_{eff}$
605: & $M_{DM}$ & t \\
606: $(M_\odot)$ & $(10^5 {\rm K})$ & $(10^{13} {\rm cm})$ & $ ({\rm gm/cm}^3)$ &
607: $({\rm gm/cm}^3)$ & $(L_\odot)$ & $(10^3 {\rm K})$
608: & (gm) & (yr) \\
609: \hline
610: 12
611: & $1.3$
612: & $4.2$
613: & $4.1 \times 10^{-7}$
614: & $1.1 \times 10^{-9}$
615: & $1.1 \times 10^5$
616: & $4.3$
617: & $2.8 \times 10^{31}$
618: & $6 \times 10^3$ \\
619: \hline
620: 50
621: & $2.7$
622: & $6.0$
623: & $6.2 \times 10^{-7}$
624: & $1.2 \times 10^{-9}$
625: & $4.2 \times 10^5$
626: & $5.0$
627: & $9.1 \times 10^{31}$
628: & $2.5 \times 10^4$ \\
629: \hline
630: 100
631: & $3.5$
632: & $7.1$
633: & $7.7 \times 10^{-7}$
634: & $1.1 \times 10^{-9}$
635: & $7.8 \times 10^5$
636: & $5.3$
637: & $1.6 \times 10^{32}$
638: & $5 \times1 0^4$ \\
639: \hline
640: 300
641: & $5.3$
642: & $9.0$
643: & $1.2 \times 10^{-6}$
644: & $8.2 \times 10^{-10}$
645: & $1.9 \times 10^6$
646: & $6.0$
647: & $3.6 \times 10^{32}$
648: & $1.5 \times 10^5$ \\
649: \hline
650: 1000
651: & $8.5$
652: & $10$
653: & $2.4 \times 10^{-6}$
654: & $4.5 \times 10^{-10}$
655: & $3.9 \times 10^6$
656: & $6.6$
657: & $7.3 \times 10^{32}$
658: & $5\times10^5$
659: \\
660: \hline
661: \hline\hline
662: \end{tabular}
663: \label{tab:ExpParam}
664: }
665: \end{center}
666: \end{table*}
667:
668: \clearpage
669:
670: \begin{figure}[t]
671: %\epsscale{0.7}
672: \plotone{f1.eps}
673: %\centerline{\includegraphics[width=0.4\textwidth]{f1.eps}}
674: \caption{Evolution of a dark star (n=1.5) as mass is accreted onto the initial
675: protostellar core of 3 M$_\odot$. The set of upper
676: (lower) solid curves correspond to the baryonic (DM) density profile (values given
677: on left axis) at different
678: masses and times. {\it Dashed lines: } luminosity $L_{\rm DM}$ integrated out to radius $r$
679: for the masses 12 and 1000 M$_\odot$, in solar units (values given on the right axis).
680: %After $ 5 \times 10^4$ ($5 \times 10^5$) years, the DS
681: %mass reaches $100 (1000) M_\odot$. The DM profile has been
682: %consistently adjusted to account for DM annihilation. As a result,
683: %the central DM density has decreased relative to an early time in its
684: %evolution.
685: %\vspace{0.1 in}
686: }
687: \end{figure}
688:
689: \end{document}
690:
691: