1: \documentclass[12pt,preprint]{aastex}
2:
3: %% manuscript produces a one-column, double-spaced document:
4:
5: %\documentclass[manuscript]{aastex}
6:
7: %% preprint2 produces a double-column, single-spaced document:
8:
9: %\documentclass[preprint2]{aastex}
10:
11: %% \documentclass[preprint2,longabstract]{aastex}
12: %\usepackage[dvips]{color}
13:
14: \newcommand{\vdag}{(v)^\dagger}
15: \newcommand{\myemail}{tnozawa@mail.sci.hokudai.ac.jp}
16:
17: \slugcomment{accepted for publication in Ap.J.}
18:
19: \shorttitle{DUST EVOLUTION IN PRIMORDIAL SNRS}
20: \shortauthors{NOZAWA et al.}
21:
22: \begin{document}
23:
24: \title{EVOLUTION OF DUST IN PRIMORDIAL SUPERNOVA REMNANTS: CAN DUST
25: GRAINS FORMED IN THE EJECTA SURVIVE AND BE INJECTED INTO THE EARLY
26: INTERSTELLAR MEDIUM?}
27:
28: \author{TAKAYA NOZAWA,\altaffilmark{1} TAKASHI KOZASA,\altaffilmark{1}
29: ASAO HABE,\altaffilmark{1} ELI DWEK,\altaffilmark{2}
30: HIDEYUKI UMEDA,\altaffilmark{3} NOZOMU TOMINAGA\altaffilmark{3},
31: KEIICHI MAEDA\altaffilmark{4} and KEN'ICHI NOMOTO\altaffilmark{3,5}}
32: %\affil{Department of Cosmosciences, Graduate School
33: %of Science, Hokkaido University, Sapporo 060-0810, Japan;
34: %tnozawa@mail.sci.hokudai.ac.jp, kozasa@mail.sci.hokudai.ac.jp,
35: %habe@astro1.sci.hokudai.ac.jp}
36: %
37: %\and
38: %
39: %\author{ELI DWEK}
40: %\affil{Laboratory for Astronomy and Solar Physics, NASA Goddard Space
41: %Flight Center, Greenbelt, MD 20771; eli.dwek@nasa.gov}
42:
43: \altaffiltext{1}{Department of Cosmosciences, Graduate School
44: of Science, Hokkaido University, Sapporo 060-0810, Japan;
45: tnozawa@mail.sci.hokudai.ac.jp}
46: \altaffiltext{2}{Laboratory for Astronomy and Solar Physics, NASA
47: Goddard Space Flight Center, Greenbelt, MD 20771}
48: \altaffiltext{3}{Department of Astronomy, School of Science, University
49: of Tokyo, Bunkyo-ku, Tokyo 113-0033, Japan}
50: \altaffiltext{4}{Max-Planck-Institut f\"ur Astrophysik,
51: Karl-Schwarzschild Strasse 1, 85741 Garching, Germany}
52: \altaffiltext{5}{Research Center for the Early Universe, School of
53: Science, University of Tokyo, Bunkyo-ku, Tokyo 113-0033, Japan}
54:
55:
56: \begin{abstract}
57:
58: We investigate the evolution of dust that formed at Population III
59: supernova (SN) explosions and its processing through the collisions with
60: the reverse shocks resulting from the interaction of the SN ejecta with
61: the ambient medium.
62: In particular, we investigate the transport of the shocked dust within
63: the SN remnant (SNR), and its effect on the chemical composition, the
64: size distribution, and the total mass of dust surviving in SNR.
65: We find that the evolution of the reverse shock, and hence its effect on
66: the processing of the dust depends on the thickness of the envelope
67: retained by the progenitor star.
68: Furthermore, the transport and survival of the dust grains depend on
69: their initial radius, $a_{\rm ini}$, and composition:
70: For Type II SNRs expanding into the interstellar medium (ISM) with a
71: density of $n_{\rm H,0}=1$ cm$^{-3}$, small grains with $a_{\rm ini} \la
72: 0.05$ $\micron$ are completely destroyed by sputtering in the postshock
73: flow, while grains with $a_{\rm ini}=$ 0.05--0.2 $\micron$ are trapped
74: into the dense shell behind the forward shock.
75: Very large grains of $a_{\rm ini} \ga 0.2$ $\micron$ are ejected into
76: the ISM without decreasing their sizes significantly.
77: We find that the total mass fraction of dust that is destroyed by the
78: reverse shock ranges from 0.2 to 1.0, depending on the energy of the
79: explosion and the density of the ambient ISM.
80: The results of our calculations have significant impact on the abundance
81: pattern of subsequent generation of stars that form in the dense shell
82: of primordial SNRs.
83:
84: \end{abstract}
85:
86: \keywords{dust, extinction --- early universe --- shock waves ---
87: supernova remnants --- supernovae: general}
88:
89:
90: \section{INTRODUCTION}
91:
92: Recent far-infrared to millimeter observations of quasars with redshifts
93: $\ga 5$ have revealed the presence of large amount of dust with masses
94: in excess of $10^8$ $M_{\odot}$ (Bertoldi et al. 2003; Priddey et al.
95: 2003; Robson et al. 2004; Beelen et al. 2006).
96: The presence of these large quantities of dust at such early epoch when
97: the universe was $\la$1 Gyr old, suggests the rapid enrichment with dust
98: that formed in the explosive ejecta of short-lived massive stars (Morgan
99: \& Edmunds 2003, Maiolino et al. 2004b; Dwek et al. 2007).
100: In addition, Maiolino et al. (2004a) have reported that the dust
101: extinction curve of the broad absorption line quasar SDSS1048+46 at
102: $z=6.2$ is quite different than those of quasars at $z<4$, suggesting
103: different dust sources and evolutionary histories.
104:
105: Dust plays a pivotal role in the interstellar processes that determine
106: the state of the interstellar medium (ISM).
107: Dust affects the thermal and chemical balance of the ISM by reprocessing
108: the radiative outputs from stars, providing photoelectrons that heat the
109: gas, and depleting the gas of refractory elements that are important
110: cooling agents of the ISM.
111: Dust also serves as a catalyst for chemical reactions, especially the
112: formation of H$_2$ molecules on the surface of dust grains (Hirashita \&
113: Ferrara 2002; Cazaux \& Spaans 2004).
114: In addition, the cooling of gas through thermal radiation from dust
115: triggers the fragmentation of star-forming cloud into low-mass gas
116: clumps of $\sim$0.1--1 $M_\odot$ even for the metallicity of
117: $10^{-6}$--$10^{-5}$ $Z_\odot$ (Omukai et al. 2005; Schneider et
118: al. 2006), although very massive stars of $\ga$100 $M_\odot$ are
119: considered to be formed up to $Z \simeq 10^{-3.5}$ $Z_\odot$ without the
120: effect of dust (Bromm et al. 2001).
121: Finally, dust obscures the nature of underlying stellar populations (e.g.,
122: Hines et al. 2006) and physical processes in the early universe.
123: Understanding the origin and the complex evolutionary history of dust is
124: therefore one of the most important goals in astrophysics.
125:
126: During the first Gyr of cosmic history, supernovae (SNe) are the only
127: possible source of interstellar dust, since low mass stars have not had
128: time to evolve off the main sequence and inject the dust that forms in
129: their quiescent outflows into the ISM.
130: Theoretical studies aimed at determining the composition and yield of
131: dust in the ejecta of primordial Type II SNe (SNe II) and
132: pair-instability SNe (PISNe) were conducted by Todini \& Ferrara (2001),
133: Nozawa et al. (2003), and Schneider et al. (2004).
134: These works have shown that even the first SNe evolving from zero-metal
135: progenitor stars can efficiently produce dust at 150--800 days after the
136: explosion.
137: The ratio of the total dust mass to the progenitor mass $M_{\rm pr}$ is
138: 0.02--0.05 for SNe II with $M_{\rm pr}=$ 12--35 $M_\odot$ and 0.15--0.3
139: for PISNe with $M_{\rm pr}=$ 140--260 $M_\odot$.
140: The composition of the newly formed dust grains are controlled by the
141: elemental composition inside the He core, and their sizes range from
142: 0.001 $\micron$ to 1 $\micron$, depending on the concentration of the
143: gas species forming dust and the time evolution of temperature and
144: density of gas.
145:
146: The results of these calculations have been applied to study the
147: high-redshift dust.
148: Maiolino et al. (2004b) demonstrated a SN origin for high-redshift dust,
149: showing that the extinction curve of the $z=6.2$ quasar SDSS1048+46 can
150: be nicely fitted by the SN II dust models from Todini \& Ferrara (2001).
151: Dust produced in the unmixed SNe II by Nozawa et al. (2003) can also
152: successfully reproduce the extinction curve of the quasar SDSS1048+46
153: by weighting the progenitor mass with the Salpeter initial mass function
154: (Hirashita et al. 2005).
155: Adopting the dust models by Nozawa et al. (2003), Nozawa et al. (2006)
156: investigated the destruction of dust in the early ISM by the
157: high-velocity shocks driven by SNe, and derived the timescale of dust
158: destruction in the early universe as a function of the explosion energy
159: of SNe and the gas density in the ISM.
160: It should be pointed out here that these studies implicitly assumed that
161: dust grains formed in SNe are injected into the ISM without their
162: composition and size distribution being reprocessed.
163:
164: However, the interaction of the SN ejecta with the surrounding medium
165: will create a reverse shock which will process the grains that condensed
166: in the He core before their injection into the ISM.
167: Once the newly formed dust grains encounter the reverse shock, they
168: acquire the high velocities relative to the gas and penetrate into the
169: hot gas created by the passage of the reverse shock and forward shock.
170: These dust grains are eroded by the kinetic sputtering and are also
171: decelerated by the drag force of the gas.
172: Small grains decelerate efficiently, and become trapped in the hot gas,
173: where they are efficiently destroyed by thermal sputtering.
174: On the other hand, large grains can maintain their high velocities, pass
175: through the shocked gas and the outwardly expanding shock front, and be
176: injected into the ISM without significant destruction.
177: The net amount and composition of the dust that is eventually returned
178: to the ISM by SNe differs substantially from the dust that was in the SN
179: ejecta shortly after its formation.
180:
181: In this paper, we study the evolution of dust formed in primordial SNe
182: II and PISNe, considering its processing through the collisions with the
183: reverse shocks and its transport within SNRs, based on the dust
184: formation calculations by Nozawa et al. (2003).
185: The questions that we pose here are what fraction of dust grains formed
186: in Population III SNe can survive the hostile circumstances within SNRs
187: and how their size distributions can be altered by sputtering in the
188: postshock flow.
189: This subject has not been fully explored to date.
190: Recently, Bianchi \& Schneider (2007) have studied the evolution of
191: newly condensed grains through the passage of the reverse shock by using
192: a semi-analytical model and have showed that the fraction of dust mass
193: survived ranges between 2 \% and 20 \% depending on the density in the
194: ISM.
195: However, they consider only the dust evolution in the nonradiative
196: phase of SNR up to about 4--8$\times 10^4$ yr from the SN explosion,
197: without taking into account the motion of dust relative to gas caused by
198: the drag force which strongly affects the destruction process and
199: evolution of dust in SNRs.
200: In the calculations, we carefully treat the dynamics and destruction of
201: dust and the time evolution of the temperature and density of the gas
202: within SNRs
203: until $\sim$$10^5$--$10^6$ yr extending over a period of the radiative
204: phase, in order to reveal how much amount of dust is finally injected
205: into the ISM or destroyed completely.
206: Although we focus on Population III SNe in this paper, this study can
207: also give great insight into the evolution of dust in Galactic SNRs.
208:
209: In \S~2, we describe the initial conditions for the evolution of SNRs,
210: the model of dust inside the He core, and the physics of dust and gas
211: within SNRs.
212: In \S~3, we present and discuss the results of calculations.
213: In \S~4, we shall discuss the effects of the hydrogen envelope on the
214: evolution of dust in SNRs.
215: As an application of the result, we investigate the abundance patterns
216: of the second-generation stars formed in the dense shell of Population
217: III SNRs in \S~5.
218: The summary is presented in \S~6.
219:
220:
221: \section{THE MODEL OF CALCULATIONS}
222:
223: \subsection{\textit{The Initial Conditions for the Evolution of SNRs}}
224:
225: The evolution of SNR is described by three characteristic parameters;
226: explosion energy, ejecta mass, and the density profile of the ambient
227: gas (Truelove \& McKee 1999).
228: In this paper, we focus on the evolution of ejecta expanding into a
229: uniform ambient medium whose elemental composition is primordial.
230: To investigate the dependence of the efficiency of dust destruction on
231: the ambient gas density, we consider three cases for the hydrogen number
232: density in the ISM; $n_{\rm H, 0} =$ 0.1, 1, and 10 cm$^{-3}$.
233: The temperature of gas $T_0$ in the ISM can be also affect the
234: evolution of SNRs, since the ambient pressure regulates the
235: deceleration of blast wave.
236: However, we have confirmed that the results of calculations are almost
237: independent of the value of $T_0$, provided that $T_0 = 10^3$--$10^5$ K.
238: Thus, we assume here $T_0 = 10^4$ K regardless of the gas density in the
239: ISM, referring to the studies showing that the radiative feedback from
240: the massive pre-SN stars can cause the ambient ISM to heat up to $T_0
241: \sim 10^4$ K (Kitayama et al. 2004; Machida et al. 2005).
242:
243: The initial conditions for the structures of density and velocity in the
244: ejecta are taken from the hydrodynamic models of Population III SNe by
245: Umeda \& Nomoto (2002).
246: We adopt six SN models; four SNe II and two PISNe.
247: The explosion energy of SNe II with $M_{\rm pr}=$ 13, 20, 25, and 30
248: $M_\odot$ is $10^{51}$ erg, and that of PISNe with $M_{\rm pr}=$ 170 and
249: 200 $M_\odot$ is $2 \times 10^{52}$ erg and $2.8 \times 10^{52}$ erg,
250: respectively.
251: It should be mentioned here that the massive metal-free stars cannot
252: lose significant mass during their lifetime due to pulsations and
253: line-driven stellar winds which are considered to be important at high
254: metallicities (Baraffe et al. 2001).
255: Although Smith \& Owocki (2006) have suggested that the mass loss of
256: very massive stars above roughly 40--50 $M_\odot$ may be possible by
257: continuum-driven winds or hydrodynamic explosions being insensitive to
258: metallicity, the mechanism to trigger such a mass loss is an open
259: question.
260: Thus, we consider that Population III PISNe as well as SNe II have
261: retained their thick hydrogen envelopes at the time of explosion.
262:
263:
264: \subsection{\textit{Dust Model inside the He core}}
265:
266: In the ejecta of SNe, dust grains can nucleate and grow only in the
267: metal-rich cooling gas, and their composition and size distribution
268: largely depend on the elemental abundance inside the He core (Kozasa et
269: al. 1989).
270: Accordingly, Nozawa et al. (2003) have calculated the dust formation in
271: Population III SNe by considering two cases for the elemental
272: composition inside the He core.
273: They found that in the unmixed ejecta with the original onion-like
274: structure, a variety of grain species condense in each layer.
275: The main grain species are C grain in carbon-rich He layer;
276: Al$_2$O$_3$, Mg$_2$SiO$_4$, and MgO grains in O-Mg-Si layer;
277: Al$_2$O$_3$, MgSiO$_3$, and SiO$_2$ grains in O-Si-Mg layer;
278: Si and FeS grains in Si-S-Fe layer;
279: Fe grain in the innermost Fe-Ni core.
280: On the other hand, oxide (Al$_2$O$_3$ and Fe$_3$O$_4$) and silicate
281: (MgSiO$_3$, Mg$_2$SiO$_4$, and SiO$_2$) grains are formed in the
282: uniformly mixed ejecta with C/O $< 1$, where the efficiency of unity
283: is assumed for the formation of CO and SiO molecules.
284:
285: Dust grains inside the He core are never processed by kinetic or thermal
286: sputtering before they hit the reverse shock, because they move with the
287: same velocities as the gas, and the gas temperature within the He core
288: is too low for dust grains to be destroyed by the thermal sputtering.
289: Therefore, as the initial condition of dust residing within the He core,
290: we adopt the size distribution, mass fraction, and spatial distribution
291: of each dust species calculated by Nozawa et al. (2003).
292: In what follows, we refer to the dust grains created in the unmixed and
293: mixed ejecta as the unmixed grain model and the mixed grain model,
294: respectively.
295:
296:
297: \subsection{\textit{Physics of Gas and Dust in SNRs}}
298:
299: The collision of the expanding SN ejecta with the surrounding ISM
300: simultaneously creates a forward shock at the interface between the
301: ejecta and the ISM, and a reverse shock that penetrates into the ejecta.
302: Once the reverse shock encounters dust grains inside the He core, the
303: dust grains are decoupled from the comoving gas to ballistically intrude
304: into the hot gas heated by the reverse shock, and then they are eroded
305: via the kinetic sputtering because of high velocities relative to
306: gas.
307: If dust grains are trapped into the postshock flow owing to the gas
308: drag, they are destroyed via the thermal sputtering caused by the
309: thermal motion of gas.
310: These dust particles are also heated by the collisions with the gas and
311: radiate the thermal emission to cool the postshock gas.
312: The rates of deceleration, erosion by sputtering, and heating of dust
313: grains depend on not only their chemical composition and size but also
314: the temperature and density of the gas in the postshock flow.
315:
316: Recently, Nozawa et al. (2006, NKH06) have calculated the dust
317: destruction in the high-velocity interstellar shocks driven by SNe in
318: the early universe, by carefully treating the dynamics, erosion, and
319: heating of dust grains, taking account of their size distribution and
320: the time evolution of the temperature and density of the gas in the
321: postshock flow.
322: The present calculations of the dust evolution within SNRs follow the
323: method described in NKH06 (see, NKH06 for details).
324: We briefly mention the outline, focusing on the difference in the
325: cooling function used in the calculations.\footnote{
326: Note that in the calculation we neglect the effect of Coulomb drag on the
327: motion of dust:
328: The ratio of the Coulomb drag force to the gas drag force is given by
329: $\sim$$\phi^2 \ln\Lambda (G_{\rm plasma}(s)/G_{\rm coll}(s))$ (Draine
330: \& Salpeter 1979), where $\phi$ is the dimensionless potential parameter,
331: $\Lambda$ is the Coulomb cutoff factor, and $s$ is defined by
332: $s^2=m w_d^2/2kT$ with $m$ the mass of gas and $w_d$ the velocity of
333: dust relative to gas.
334: $G_{\rm plasma}/G_{\rm coll} < 1$ for $0< s < \infty$ (Draine \&
335: Salpeter 1979), $\ln\Lambda \sim$30--40 in SNRs (Dwek \& Arendt 1992),
336: and $\phi$ is evaluated to be $\sim$$10^5/T$ for the gas temperatrure of
337: $T \ga 10^{5.5}$ K (McKee et al. 1987).
338: Thus, the Coulomb drag is negligible compared with the gas drag for
339: $T \ga 10^6$ K, otherwise it can play important role in the deceleration
340: of dust.
341: However, the erosion rate of dust grains by thermal sputtering quickly
342: decreases at $T \la 10^6$ K (NKH06).
343: As a result, the Coulomb drag does not significantly affect the motion
344: and destruction of dust grains considered in this paper.}
345:
346: We assume that the spherically symmetric ejecta collides with the ISM
347: in 10 yr after the explosion.
348: With the initial conditions described in \S~2.1, the time evolution of
349: the gas in SNRs is numerically solved with the flux-splitting method
350: (van Albada, et al. 1982; Mair et al. 1988).
351: In the calculations, we include three processes of the radiative cooling
352: in the equation of the conservation of energy.
353: The first is the thermal emission from dust collisionally heated in the
354: postshock flow, and the second is the inverse Compton cooling, whose
355: rate is calculated at the redshift of $z=20$.
356: The third is the cooling of gas by the atomic process.
357: For the gas with the primordial composition in the hydrogen envelope and
358: behind the forward shock, we adopt the atomic cooling function for the
359: zero metal case given by Sutherland \& Dopita (1993) that is limited to
360: the gas temperature of $T \ge 10^4$ K.
361: In order to calculate the evolution of the gas in the dense shell
362: appeared at the later phase of SNRs, we extrapolate the cooling rate of
363: gas at $T < 10^4$ K as follows.
364: Referring to Machida et al. (2005), the atomic cooling rate at $T <
365: 10^4$ K is approximately proportional to $T^4$; the cooling rate
366: decreases from $\Lambda^{\rm gas} \simeq 10^{-23}$ erg cm$^{-3}$
367: s$^{-1}$ at $T = 10^4$ K to $\Lambda^{\rm gas} \simeq 10^{-27}$ erg
368: cm$^{-3}$ s$^{-1}$ at $T=10^3$ K.
369: Therefore, we simply evaluate the atomic cooling rate as
370: $\Lambda^{\rm gas} = T^4_4 \Lambda^{\rm gas}_4$, where $T_4$ is the gas
371: temperature in units of $10^4$ K and $\Lambda^{\rm gas}_4$ is the atomic
372: cooling rate at $T=10^4$ K.
373: For the metal-rich He core, we assume that the gas is only composed of
374: oxygen which is the most abundant gas species inside the He core, and
375: employ the cooling functions from Smith et al. (2001) for $T \ge 10^5$ K
376: and Raymond \& Smith (1977) for $T < 10^5$ K.
377: In the calculations, we ignore the contribution of cooling by metal ions
378: released from dust by sputtering for simplicity.
379:
380: The dynamics and destruction of dust after colliding with the reverse
381: shocks are calculated as follows.
382: By treating dust as a test particle and ignoring the effect of charge on
383: dust grains, the deceleration rate of dust due to the gas drag is
384: calculated for each size of the dust to evaluate the velocity relative
385: to gas and the position.
386: Then we calculate the dust destruction by sputtering and the heating by
387: collisions with the gas, using the relative velocity and the temperature
388: and density of gas at the position.
389: The sputtering yield of each dust species is calculated with the
390: universal relation derived by NKH06.
391: Dust grains are considered to be completely destroyed when their sizes
392: become smaller than the radius of the nominal monomer of condensate.
393: The cooling of gas through thermal emission of dust is calculated by
394: balancing the heating of dust resulting from the collisions with
395: electrons.
396: The calculations are performed by the truncation time at which the
397: forward shock velocity decelerates below 20 km s$^{-1}$.
398:
399:
400: \section{RESULTS}
401:
402: In this section, we present the results of the calculations of dust
403: evolution within primordial SNRs.
404: In \S~3.1, we demonstrate the time evolution of temperature and density
405: of gas in the SNR, and in \S~3.2, we show the transport and destruction
406: of dust grains within SNRs.
407: In \S~3.3, we provide the resulting size distribution of survived dust
408: and elucidate the dependence of the efficiency of dust destruction on
409: the progenitor mass and the gas density in the ISM.
410:
411:
412: \subsection{\textit{Time Evolution of the Gas in the SNRs}}
413:
414: Figure 1 shows the time evolution of the density (Fig. 1a) and
415: temperature (Fig. 1b) of gas by $2 \times 10^4$ yr in the SNR generated
416: from the explosion of star with $M_{\rm pr} = 20$ $M_\odot$ and
417: expanding into the ISM with $n_{\rm H,0}=1$ cm$^{-3}$.
418: The unmixed grain model is taken as the model of dust inside the He
419: core, and the cooling of gas by dust is taken into account.
420: In what follows, we refer to this model as the standard model.
421:
422: The forward shock resulting from the interaction of the ambient gas with
423: the ejecta of the SN is specified by the steep rise of the gas
424: temperature and the increase of the gas density by $\sim$4 times that in
425: the ISM.
426: The downward arrows in Figure 1a show the positions of the forward
427: shock.
428: We can also identify the formation of the reverse shock from high
429: temperature of gas heated by the compression of the ejecta, and indicate
430: its position for each time by the downward arrow in Figure 1b.
431:
432: The position $R_{\rm rs}$ (Fig. 2a) and velocity $V_{\rm rs}$ (Fig. 2b)
433: of this reverse shock as a function of time are depicted by the thick
434: solid curves in Figure 2.
435: The reverse shock is decelerated by the shocked ambient medium while it
436: initially expands outward, and then returns back at a distance of
437: $\sim$5 pc ($1.5 \times 10^{19}$ cm) with the velocity of a few 100 km
438: s$^{-1}$.
439: The trajectory of the reverse shock is affected by the detail structure
440: of density in the ejecta.
441: The collision with the locally high-density gas inside the He core
442: causes the reverse shock to move outward again at 5200 yr.
443: After $1.1 \times 10^4$ yr, the reverse shock goes inward through the He
444: core with increasing its velocity up to $>$1000 km s$^{-1}$.
445: Hence, dust grains crossing the reverse shock acquire the different
446: velocities relative to gas, depending on the time of collision with the
447: reverse shock, and are efficiently eroded by the kinetic sputtering
448: if the relative velocity is $\sim$500--1300 km s$^{-1}$.
449:
450: It can be seen from Figure 1 that the temperature of the gas in the
451: region between the forward and reverse shocks is higher than $10^6$ K.
452: Thus, dust grains staying in this region are subject to the thermal
453: sputtering.
454: However, the erosion rate of dust decreases as the SNR evolves because
455: the density decreases with time.
456: Around the truncation time ($\simeq$8 $\times 10^5$ yr for the standard
457: model), the gas density within the SNR is more than 100 times lower than
458: that in the ISM, and the gas temperature becomes low ($\sim$several
459: times $10^5$ K) enough for dust grains not to be sputtered efficiently,
460: and thus the destruction of dust via the sputtering is extremely
461: inefficient.
462:
463: The influence of the cooling by dust on the SNR evolution is clarified
464: by comparing the results of calculations with and without the cooling by
465: dust emission.
466: The evolution of the reverse shock without the dust cooling for
467: $M_{\rm pr}=20$ $M_\odot$ and $n_{\rm H,0} = 1$ cm$^{-3}$ is shown by
468: the thick dashed curves in Figure 2, where the cooling of gas by dust
469: can cause the velocity of the reverse shock to be reduced by $\sim$10
470: \%.
471: The thin curves in Figure 2 show the evolution of the reverse shock
472: penetrating into the ejecta of PISN with $M_{\rm pr}=170$ $M_\odot$
473: calculated for $n_{\rm H,0} = 1$ cm$^{-3}$.
474: Because the thermal emission from dust increases with increasing the
475: dust mass, the effects of cooling by dust are significant for the
476: remnants of PISNe, where the mass of dust formed in the ejecta is a few
477: tens $M_\odot$.
478: As we can see from the figure, the velocity of the reverse shock
479: including dust cooling (\textit{thin solid line}) decreases to 0.6 times
480: that not including the cooling (\textit{thin dashed line}) at $\ga$
481: 7000 yr.
482: In addition, the dust cooling decreases the gas temperature by $\sim$20
483: \%, compared with that calculated without dust cooling.
484: However, it should be noted that the efficiency of dust destruction is
485: little affected by the dust cooling, because most of dust grains are
486: predominantly destroyed by the thermal sputtering and the erosion rate
487: is not sensitive to the gas temperature as long as $T > 10^6$ K (NKH06).
488: Note that the cooling by oxygen line has no influence on the results,
489: since the temperature within the He core during the passage of the
490: reverse shock is above $10^7$ K where the dominant cooling process of
491: gas is free-free emission.
492:
493:
494: \subsection{\textit{Transport and Destruction of Dust in SNR}}
495:
496: The time evolutions of the positions (Fig. 3a) and sizes (Fig. 3b) of
497: dust grains within the SNR for the standard model are given in Figure 3.
498: In Figure 3a, the trajectories of the forward and reverse shocks are
499: also depicted by the thick solid curves, along with the position of the
500: surface of the He core.
501: Among nine dust species in the unmixed grain model, are shown C,
502: Mg$_2$SiO$_4$, and Fe grains with the initial sizes of 0.01 $\micron$
503: (\textit{dotted lines}), 0.1 $\micron$ (\textit{solid lines}), and 1
504: $\micron$ (\textit{dashed lines}), respectively.
505: Each grain species initially moves coupling with the gas with the
506: velocity of $\sim$1300, $\sim$900, and $\sim$400 km s$^{-1}$ for C,
507: Mg$_2$SiO$_4$, and Fe grains, respectively, and collides with the
508: reverse shock at 3650 yr for C grains formed in the outermost He core,
509: 6300 yr for Mg$_2$SiO$_4$ grains in the oxygen-rich layer, and 13000 yr
510: for Fe grains condensed in the innermost He core.
511:
512: The collision time of the reverse shocks with dust grains depends on the
513: initial velocity and position of dust, the thickness of the hydrogen
514: envelope, and the density in the ISM.
515: Figure 4 gives the collision time $t_{\rm coll}$ of the reverse shock
516: with the He core for different progenitor mass and gas density in the
517: ISM.
518: We can see that the collision time is shorter for PISNe than SNe II, in
519: spite of the fact that PISNe have the thicker hydrogen envelopes.
520: This reason is that the gas velocities ($\sim$2000--3000 km s$^{-1}$) at
521: the outermost He core for PISNe with the explosion energies higher than
522: 10$^{52}$ erg are a few times higher than those for SNe II.
523: For $n_{\rm H,0} =$ 0.1--10 cm$^{-3}$, the collision times are
524: $\simeq$10$^3$--10$^4$ yr and decrease with increasing the ambient gas
525: density.
526: Note that this result is not consistent with the observations of the Cas
527: A SNR, where thermal emission from dust heated by the reverse shock has
528: already detected at $\simeq$330 yr after the explosion (Ennis et al.
529: 2006 and references therein).
530: This is because the progenitor of the Cas A is believed to have lost the
531: considerable hydrogen envelope during their evolution (Young et al.
532: 2006), in contrast to the Population III SNe considered in this paper.
533: We shall discuss the effect of the hydrogen envelope on the evolution of
534: dust in the SNR in \S~4.
535:
536: The fates of dust grains within SNRs heavily depend on their initial
537: sizes $a_{\rm ini}$ as well as the chemical composition reflecting the
538: difference in the sputtering yield and bulk density, as shown in Figure
539: 3.
540: For the standard model, the relatively small grains with $a_{\rm
541: ini}=0.01$ $\micron$ are efficiently decelerated due to the gas drag and
542: are fully trapped into the hot plasma to be completely destroyed by the
543: thermal sputtering.
544: Note that the gas drag on grains with small radii is more efficient than
545: that on larger grains, because the deceleration rate of grain is
546: inversely proportional to its size (NKH06).
547: Actually, the grains with the initial sizes less than 0.05 $\micron$
548: are trapped into the hot gas between the forward and reverse shocks and
549: continue to be eroded by the thermal sputtering even at over $10^5$ yr
550: until they are completely destroyed.
551: Larger grains of $a_{\rm ini} = 0.1$ $\micron$ undergo the kinetic
552: and thermal sputtering while streaming in the hot gas.
553: Thanks to the high bulk density, Fe grains with $a_{\rm ini}=0.1$
554: $\micron$ are injected into the ISM, reducing the size by 52 \%.
555: On the other hand, C and Mg$_2$SiO$_4$ grains with $a_{\rm ini}=0.1$
556: $\micron$ are trapped and eroded by the thermal sputtering in the denser
557: region behind the forward shock, and their surface layers whose
558: thicknesses are 43 and 69 \% of their initial sizes are eroded,
559: respectively, until $2 \times 10^5$ yr when the SNR enters into the
560: radiative phase and the dense SN shell is formed behind the forward
561: shock.
562: These dust grains remain in the dense shell without further processing
563: because the gas cools down quickly below $10^6$ K.
564: Thus, the decrease of sizes of these grains is truncated at a given size,
565: and the erosion of dust with $a_{\rm ini} =$ 0.05--0.2 $\micron$ results
566: in the final size of 0.001--0.1 $\micron$, depending on their initial
567: sizes.
568: For 1 $\micron$-sized C, Mg$_2$SiO$_4$, and Fe grains, the kinetic
569: sputtering reduces their sizes by 0.7, 6, and 8 \%, respectively.
570: Note that large grains with $a_{\rm ini} \ga 0.2$ $\micron$ can go across
571: even the forward shock and be injected into the ISM, because the
572: deceleration due to the gas drag is very inefficient.
573:
574: We should mention here that the degree of the erosion of dust is a
575: complex function of the initial position and initial size of the dust as
576: well as the sputtering yield.
577: Fe grains formed in the innermost He core and travelling through the
578: oxygen-rich gas undergo the efficient erosion, since the sputtering
579: yield by an oxygen ion is $\sim$50 times larger than that by a proton.
580: In contrast, the degree of the erosion of C grains in the outermost He
581: core is relatively small because they can quickly escape from the He
582: core after the collision with the reverse shock and also have the
583: sputtering yield lower than those of other dust species.
584: Furthermore, without being decelerated efficiently, the larger grains
585: with the high velocities relative to gas are more efficiently eroded by
586: sputtering in the relatively dense region near the forward shock front.
587: However, the grains with the sizes larger than a given size can evade
588: the erosion in the shocked hot gas within SNRs and are expelled into the
589: ISM.
590:
591: Although the processing of dust in the ISM is not the main subject of
592: this paper, here we shall simply show the processing of dust injected
593: into the ISM as a consequence of dust evolution in SNRs.
594: Dust grains injected into the ISM can be consumed through the kinetic
595: sputtering because of the difference in velocity between the ambient
596: cool gas and the dust grains, while the grains can be also decelerated
597: by the direct collisions with the gas.
598: In this case, the ratio of the final radius $a_{\rm fin}$ to the initial
599: size $a_{\rm ini}$ of dust is dependent only on the escape velocity
600: $w_0$ defined as the velocity with which dust is injected into the ISM
601: passing through the forward shock, and is given by
602: %
603: \begin{eqnarray}
604: \frac{a_{\rm fin}}{a_{\rm ini}} = \exp \left[ - \left(
605: \frac{2 m_{\rm sp}}{3 \mu_g} \right) \int^{w_0}_{w_{\rm f}}
606: \sum_i A_i Y^0_i(w) \frac{dw}{w} \right],
607: \end{eqnarray}
608: %
609: where $m_{\rm sp}$ is the average atomic mass of the elements sputtered
610: from the grain, $\mu_g$ is the mean molecular weight of the gas, and
611: $Y^0_i(w)$ is the sputtering yield at normal incidence by gas species
612: $i$ whose number abundance is $A_i$.
613:
614: Figure 5 shows the results calculated by Equation (1) for the primordial
615: gas composition as a function of the escape velocity.
616: In the calculations, the final relative velocity $w_{\rm fin}$ is taken
617: as 10 km s$^{-1}$, which is small enough for dust grains not to be
618: eroded by the kinetic sputtering.
619: Also we assume that the escape velocity equals to the initial velocity
620: inside the He core, since very large grains are ejected to the ISM with
621: high velocities, not being decelerated efficiently.
622: Hence, the calculated $a_{\rm fin}$ gives the lower limit of the final
623: size realized in the ISM.
624: The final size acquired by each dust species is different, depending on
625: the sputtering yield $Y^0_i$ and the average atomic mass $m_{\rm sp}$,
626: and decreases with increasing the escape velocity.
627: The ratio of the final size to the initial size is $\simeq$0.8 for C
628: grains with $w_0 \sim 1300$ km s$^{-1}$, $\simeq$0.5 for Mg$_2$SiO$_4$
629: grains with $w_0 \sim 900$ km s$^{-1}$, and $\simeq$0.5 for Fe grains
630: with $w_0 \sim 400$ km s$^{-1}$.
631: Thus, the sizes of very large grains supplied from SNe are decreased to
632: 0.5--0.8 times those at the time of the ejection, but are not completely
633: destroyed in the ISM.
634: For PISNe with the explosion energies higher than $10^{52}$ erg, the
635: sizes of Fe grains with the initial velocities of $\simeq$1000 km
636: s$^{-1}$ decrease by 70 \% in the ISM, although the sizes of C and
637: Mg$_2$SiO$_4$ grains whose initial velocities are in the range of
638: 2000--3000 km s$^{-1}$ are not significantly different from those
639: calculated for the standard model.
640: Note that dust grains in the ISM are also processed by the high-velocity
641: interstellar shocks driven by the ambient SNe (NKH06), which is the
642: major mechanism of the dust destruction in the ISM.
643:
644:
645: \subsection{\textit{Efficiency of Dust Destruction}}
646:
647: The results described in \S~3.2 imply that the size distribution of
648: survived dust is greatly deficient in small-sized grains, compared with
649: that at the time of dust formation.
650: In Figure 6, we present the initial size distribution at the time of
651: dust formation (Fig. 6a) and the size distribution of dust at the
652: truncation time (Fig. 6b) for the standard model.
653: The comparison of these figures clearly indicates that grains with
654: the radii below a few tens \AA~are missing for almost all dust species.
655: In particular, Al$_2$O$_3$ grains are completely destroyed because their
656: initial sizes are smaller than 0.02 $\micron$.
657: We can also see that the sizes of the relatively large grains are
658: shifted to small sizes due to the erosion by sputtering.
659: It should be pointed out here that the size distribution of survived
660: dust is different from that by Bianchi \& Schneider (2007):
661: They have found that the final size distribution show a flattening
662: towards smaller sizes without any abrupt truncation.
663: The main reason of this difference is considered as follows.
664: Bianchi \& Schneider (2007) assumed that newly formed grains remain
665: confined and trapped at their initial positions in the ejecta.
666: Therefore, dust grains remaining at their initial positions cease to
667: be eroded by sputtering on the timescale of 4--8$\times 10^4$ yr due to
668: the decrease of the gas density caused by expansion.
669: As shown in \S~3.2, our results show that even small grains of $a_{\rm
670: ini} \le 0.05$ $\micron$ penetrate into the hot plasma between the
671: forward and reverse shocks, and are trapped and completely destroyed in
672: the hot gas with the relatively high density.
673: On the other hand, the erosion of grains with the radii of 0.05--0.2
674: $\micron$ produces the grains with the final sizes of 0.001--0.1
675: $\micron$ remaining in the dense shell.
676:
677: The critical size below which dust is destroyed is sensitive to the gas
678: density in the ambient medium.
679: Although the critical size weakly depends on the dust species and their
680: initial positions, we can roughly estimate the average critical size for
681: different gas density in the ISM.
682: For $n_{\rm H,0}=0.1$ cm$^{-3}$, the upper limit of the initial size of
683: dust completely destroyed is $\sim$0.01 $\micron$, and the lower limit
684: of that ejected to the ISM is $\sim$0.03 $\micron$.
685: For $n_{\rm H,0}=10$ cm$^{-3}$, the grains with $a_{\rm ini} \la 0.2$
686: $\micron$ are destroyed, and the grains with $a_{\rm ini} \ga 0.5$
687: $\micron$ are injected in the ISM.
688: Note that the above critical sizes are true for the grain species except
689: for C grains.
690: The critical size of C grain is 0.006, 0.02, and 0.07 $\micron$ for
691: $n_{\rm H,0}=$0.1, 1, and 10 cm$^{-3}$, respectively, and is a few times
692: smaller than other dust species, because C grains are located at the
693: outermost He core and also have the lower erosion rate by sputtering.
694: Note that the initial sizes of Fe grains injected into the ISM are
695: smaller than those of others by a factor of $\sim$2 because of the high
696: bulk density, and are $\ga$0.02, 0.1, and 0.25 $\micron$ for
697: $n_{\rm H,0}=$0.1, 1, and 10 cm$^{-3}$, respectively.
698:
699: On the other hand, the critical size for each grain species is almost
700: independent of the progenitor mass as long as the explosion energy of SN
701: is the same, because the time evolution of temperature and density of
702: gas within SNRs is similar.
703: For PISNe with the explosion energies higher than $10^{52}$ erg and more
704: massive hydrogen envelopes, the critical size increases approximately by
705: a factor of three, compared with that for SNe II.
706: However, the initial size of dust ejected to the ISM is less than 2
707: times that for SNe II, because the high initial velocity inside the He
708: core makes the dust grains easily escape from SNRs.
709: These results lead to the conclusion that the large-sized grains
710: dominate the mass of dust injected from SNe into the ISM.
711:
712: Tables 1 and 2 summarize the mass fraction of dust destroyed
713: $\epsilon_{\rm dest}$, piled up in the dense shell $\epsilon_{\rm
714: shell}$, and ejected to the ISM $\epsilon_{\rm eject}$ for the unmixed
715: and mixed grain models, respectively.
716: In Figures 7a and 7b, we also present the total mass of survived dust
717: for both grain models, respectively, along with their initial total
718: mass.
719: We can first see that the mass of dust destroyed increases with
720: increasing the surrounding gas density.
721: In particular, for $n_{\rm H, 0}=10$ cm$^{-3}$, all or almost all
722: ($\ga$85 \%) of dust grains formed in the ejecta are destroyed, and the
723: mass of survived dust is less than $0.1$ $M_\odot$ for all models
724: considered in this paper.
725: This reason is as follows; the erosion of dust takes place in the hot
726: gas between the forward and reverse shocks, whose density increases with
727: the ambient gas density. Therefore, the higher ISM gas density leads to
728: the efficient erosion and deceleration of dust through the more frequent
729: collisions with the hot gas.
730: Next, we find that the mass fraction of dust destroyed is generally
731: higher for the mixed grain model than the unmixed grain model;
732: for SNe II with $n_{\rm H, 0}=$ 1 (0.1) cm$^{-3}$, $\epsilon_{\rm dest}
733: =$ 0.5--0.8 (0.2--0.4) for the unmixed grain model, while
734: $\epsilon_{\rm dest} =$ 0.9--0.99 (0.57--0.78) for the mixed grain
735: model.
736: This reflects the fact that the mixed grain model lacks the grains
737: larger than 0.05 (0.01) $\micron$ in comparison with the unmixed grain
738: model.
739: Finally, it can be seen that dust grains formed in PISNe are more
740: efficiently destroyed than those in SNe II, since the newly formed dust
741: grains are dominated by the small-sized grains and the critical size is
742: much larger.
743: Thus, for $n_{\rm H, 0}=1$ cm$^{-3}$, the mass of dust survived in
744: PISNe is 0.1--1 $M_\odot$ for both grain models and is almost
745: the same as that (0.07--0.5 $M_\odot$) in the SNe II, though the mass of
746: dust formed in PISNe is a few tens times higher than that in SNe II.
747: Note that the dust destruction by reverse shocks is more effective
748: than that by high-velocity interstellar shocks for the dust grains with
749: the same size distribution (NKH06).
750:
751: The destruction efficiency defined as the ratio of the mass of dust
752: destroyed to the initial dust mass is given in Tables 3 and 4 for each
753: dust species in the unmixed and mixed grain model, respectively.
754: As mentioned above, the efficiency of dust destruction is greatly
755: influenced by what fraction of the initial dust mass is occupied by the
756: sizes larger than the critical size.
757: Thus, the destruction efficiency for a given dust species is very
758: sensitive to the initial size distribution.
759: Note that it is difficult to find the clear dependence of the
760: destruction efficiency of each dust species on the progenitor mass
761: because the size distribution of dust at the time of dust formation is
762: different from model to model.
763: However, for example, the efficiency of destruction of Al$_2$O$_3$ grain
764: with $a_{\rm ini} \la 0.02$ $\micron$ is $\sim$1 for almost all cases
765: considered in this paper.
766: For the unmixed grain model, FeS and MgSiO$_3$ grains are also
767: predominantly destroyed.
768: On the other hand, Fe, Si, and C grains for which most of the mass is
769: locked into the grains larger than 0.05 $\micron$ have the relatively
770: small destruction efficiencies.
771: For the mixed grain model, SiO$_2$ grains are the main dust species that
772: are left in SNRs and/or are injected into the ISM.
773: Therefore, we conclude that the chemical composition, size distribution,
774: and amount of dust grains supplied from SNe to the ISM are quite
775: different from those at the time of dust formation.
776:
777: Here we shall discuss the dependence of the metallicity on the present
778: results of calculations.
779: First, it should be pointed out that the species of dust formed in the
780: ejecta of SNe II and their size distributions are not sensitive to the
781: metallicity of progenitor stars (Todini \& Ferrara 2001; Nozawa 2003).
782: Also the cooling function of gas for $Z \le 10^{-3}$ $Z_{\odot}$ is
783: independent of the metallicity (Sutherland \& Dopita 1993).
784: Thus, the results of calculations presented in this paper can be
785: directly applied to the evolution of dust in the ejecta of SNe II
786: expanding into the ambient medium whose metallicity is less than
787: 10$^{-3}$ $Z_{\odot}$.
788: The increase of the metallicity in the ambient medium to the solar value
789: greatly enhances the cooling of gas behind the forward shock and causes
790: the time of the transition from nonradiative phase to radiative phase to
791: be reduced to less than half of that for the zero-metallicity case.
792: Nevertheless, the destruction efficiency of each grain species for
793: $Z = Z_\odot$ decreases at most by 15 \% of that for $Z=0$.
794: Therefore, the results of present study could be useful for evaluating
795: the dust evolution in SNRs generated from SNe II, regardless of the
796: initial metallicity of progenitor stars and ambient medium.
797: However, as is demonstrated in the next section, the thickness of
798: hydrogen envelope strongly affects the motion and destruction of dust
799: within SNRs.
800:
801: \section{\textit{THE EFFECT OF THE HYDROGEN ENVELOPE ON THE EVOLUTION OF
802: DUST IN SNR}}
803:
804: As mentioned in \S~3.2, for the SNe with the thick hydrogen envelopes,
805: it takes at least more than 1000 yr for the reverse shocks to collide
806: with the dust condensed inside the He core.
807: On the other hand, the infrared observations of the Cas A SNR (Dwek et
808: al. 1987; Lagage et al. 1996; Arendt et al. 1999; Douvion et al. 2001;
809: Hines et al. 2004; Ennis et al. 2006) have revealed the thermal emission
810: from warm dust formed in the ejecta at $\simeq$330 yr after the
811: explosion.
812: This difference of the time at which dust grains inside the He core are
813: swept up by the reverse shocks is attribute to the difference in the
814: thickness of the hydrogen envelope of SNe.
815: Therefore, in this section, we investigate the effect of the hydrogen
816: envelope on the evolution of dust in the SNR.
817:
818: In order to illustrate the evolution of dust in the SNR generated from
819: the SN whose hydrogen envelope is very thin, we adopt the model of
820: ejecta with $M_{\rm pr} = 20$ $M_\odot$ from Umeda \& Nomoto (2002) and
821: modify it by artificially reducing the mass of the hydrogen envelope
822: from 13 $M_\odot$ to 0.7 $M_\odot$ at the time of explosion.
823: Keeping the structure of the density, we simply scale up the gas
824: velocity so that the explosion energy can equal to 10$^{51}$ erg.
825: We also assume the dust inside the He core is the same as that formed in
826: primordial SN II with $M_{\rm pr} = 20$ $M_\odot$.
827:
828: Figure 8 illustrates the trajectories (Fig. 8a) and time evolutions of
829: sizes (Fig. 8b) of C, Mg$_2$SiO$_4$, and Fe grains with $a_{\rm ini} =$
830: 0.01 and 0.1 $\micron$ in the SNR calculated for $n_{\rm H,0}=$ 1
831: cm$^{-3}$.
832: As can be expected, because the hydrogen envelope is very thin and the
833: initial velocities of dust grains are very high (2000--5000 km s$^{-1}$
834: depending on their initial positions), the dust grains inside the He
835: core collide with the reverse shock at much earlier time;
836: 120 yr for C grains, 570 yr for Mg$_2$SiO$_4$ grains, and 1300 yr for Fe
837: grains.
838: Contrary to SNe II with the thick hydrogen envelopes, even the grains
839: with $a_{\rm ini} = 0.1$ $\micron$ can be ejected to the ISM without
840: decreasing their sizes significantly;
841: C and Mg$_2$SiO$_4$ grains with $a_{\rm ini} = 0.1$ $\micron$ reduce
842: their sizes only by less than 6 \%, while the sizes of Fe grains with
843: $a_{\rm ini} =0.1$ $\micron$ decrease by 18 \% because the time staying
844: in the hot gas is long, compared with C and Mg$_2$SiO$_4$ grains.
845: The grains with $a_{\rm ini} = 0.01$ $\micron$ are trapped into the hot
846: gas and are completely destroyed.
847: In this case, the critical size below which dust is completely destroyed
848: is $\sim$0.04 $\micron$, and the lower limit of the initial size of dust
849: supplied to the ISM is $\sim$0.04 $\micron$ which is five times smaller
850: than that for a SN II.
851: Therefore, the thin hydrogen envelope as well as the high initial
852: velocity of dust causes almost all survived dust grains to be injected
853: into the ISM without being trapped into the gas within the SNR.
854: The mass fraction of dust destroyed is 0.38, which is lower than 0.75
855: for the standard model, and thus more dust grains are supplied to the
856: ISM.
857: The result of calculation shows that the fates of dust grains formed in
858: the ejecta strongly depend on the thickness of the hydrogen envelope.
859:
860: However, we adopt the model of dust formed in a SN II and also consider
861: the uniform ambient medium.
862: The significant mass loss of massive stars during their evolution
863: results in the circumstellar medium that is not uniform and homogeneous.
864: In addition, the time evolution of the temperature and density of gas in
865: the ejecta of SNe without the hydrogen envelopes such as Type Ib/c SNe
866: is expected to be different from that in SNe II, which influences the
867: dust formation in the ejecta.
868: Therefore, we must clarify the size distribution and amount of dust
869: grains formed in the ejecta to investigate the dependence of the
870: evolution of dust on the type of SNe.
871: This subject will be reported in the forthcoming paper.
872:
873:
874: \section{\textit{ELEMENTAL ABUNDANCES OF THE SECOND-GENERATION STARS}}
875:
876: Some extremely metal-poor (EMP) stars with [Fe/H] $\la -2.5$ discovered
877: in the Galactic halo have peculiar abundance patterns showing the modest
878: or large (1--100 times) enhancements of C, N, O, Mg, and Si relative to
879: solar (Venkatesan et al. 2006 and references therein).
880: In particular, two hyper metal-poor (HMP) stars with [Fe/H] $\la -5$, HE
881: 0107-5240 ([Fe/H] $=-5.2$, Christlieb et al. 2002) and HE 1327-2326
882: ([Fe/H] $=-5.4$, Frebel et al. 2005) show extreme ($\ga$10$^2$ times)
883: overabundances of C, N, and O relative to iron.
884: Because the elemental composition of HMP stars expected to be the very
885: early generation may strongly reflect the nucleosynthesis in Population
886: III stars, several scenarios have been proposed to explain the origin of
887: elemental abundances in low-mass HMP stars;
888: the mixing fall-back in a core-collapse SN (Umeda \& Nomoto 2005;
889: Iwamoto et al. 2005), the nucleosynthesis and mass transfer in a
890: first-generation binary star (Suda et al. 2004; Komiya et al. 2006),
891: the pollution by the gas with heavy elements in the ISM (Shigeyama et
892: al. 2003), and the combinations of these scenarios described above
893: (Christlieb et al. 2004).
894:
895: Recently, Venkatesan et al. (2006) have proposed that EMP stars are the
896: second-generation stars formed in the dense shell of primordial SNRs and
897: their peculiar abundances can be reproduced by the segregated transport
898: of newly formed dust decoupled from the metal-rich gas in SNe.
899: Adopting the dust models by Todini \& Ferrara (2001) and Schneider et
900: al. (2004), they calculated the sputtering and the transport of dust
901: driven by the UV radiation field from the stellar cluster within a SNR.
902: They have conclude that the progenitor mass range of 10--150 $M_\odot$
903: qualitatively explains the enhancement of the elements composing dust
904: grains (C, O, Mg, and Si) in EMP stars, though they did not
905: compare their results with the abundance data on these metal-poor stars.
906:
907: The results of the transportation and destruction of dust given in
908: \S~3.2 show that all dust grains remaining in primordial SNRs without
909: being completely destroyed by sputtering can be accumulated in the dense
910: SN shell at $\sim$10$^5$--10$^6$ yr after the explosion.
911: Thus, it is considered that the formation of the second-generation stars
912: with solar mass scales could be possible in the shell contaminated with
913: dust grains, as is investigated by Schneider et al. (2006).
914: In this case, we can expect that the elemental compositions of dust
915: grains piled up into the dense shell reflect the metal abundance
916: patterns of the second-generation stars.
917:
918: The iron-bearing dust species including the most important element Fe
919: are Fe and FeS grains for the unmixed grain model, and Fe$_3$O$_4$ grain
920: for the mixed grain model.
921: Note that C grains cannot be formed in the mixed ejecta
922: because the ejecta is oxygen-rich, and
923: Fe$_3$O$_4$ grains with $a_{\rm ini} \la 0.05$ $\micron$ are dominantly
924: destroyed in the postshock flow and are rarely accumulated in the SN
925: shell for $n_{\rm H,0}=$ 1 and 10 cm$^{-3}$.
926: Hence, we show only the results calculated for the unmixed grain model.
927: In addition, Fe and FeS grains are not significantly piled up in the
928: shell of PISN remnants, since most of them are completely destroyed or
929: are injected into the ISM.
930: Thus, we focus on the results for Type II SNRs.
931:
932: The abundances of C, O, Mg, and Si relative to Fe in the dense shell
933: are summarized in Tables 5, where we neglect the contribution of metal
934: atoms sputtered from the grains crossing the dense shell
935: because the largest-sized grains are only eroded by less than 1\% of their
936: radii.
937: The hydrogen mass in the dense SN shell $M^{\rm H}_{\rm shell}$ at the
938: truncation time is given in Table 1.
939: It can be seen from Table 5 that all models considered here predict the
940: value of [Fe/H] less than $-4$, and most of them exhibit the value of
941: [Fe/H] ranging from $-6$ to $-5$.
942: Thus, two HMP stars are considered to be formed in the dense shell of
943: primordial SNRs.
944: Among the 9 models with $-6 \le$ [Fe/H] $\le -5$, the abundances of Mg
945: and Si are in the range of $-1.2 \le$ [Mg/Fe] $\le 1.2$ and
946: $-0.6 \le$ [Si/Fe] $\le 2.7$, respectively, and 8 models can produce
947: the overabundances of Mg and/or Si.
948: This result suggests that the dust formed in the unmixed ejecta of SNe
949: II can be responsible for the peculiar abundance patterns of Mg and Si
950: in HMP stars.
951: Although a few models (20 $M_\odot$ and 30 $M_\odot$ for $n_{\rm
952: H,0} =$ 1 cm$^{-3}$ and 25 $M_\odot$ for $n_{\rm H,0} =$ 0.1 cm$^{-3}$)
953: result in the overabundances of C and O as well as Mg and Si, [C/Fe] and
954: [O/Fe] are limited to $<$1.6 and $<$0.6, respectively.
955:
956: For comparison, we apply the present calculations to the models of dust
957: formation in the SN II with $M_{\rm pr} =$ 22 $M_\odot$ by Todini \&
958: Ferrara (2001) and in PISN with $M_{\rm pr} =$ 195 $M_\odot$ by
959: Schneider et al. (2004).
960: Because the sizes of the dust grains calculated by them are considerably
961: small ($\la$0.04 $\micron$) except for C grain, the grain species that can
962: survive the destruction through the collisions with the reverse shocks
963: is only C grain for the ambient gas density of $n_{\rm H,0}=$ 0.1--10
964: cm$^{-3}$.
965: Thus, their dust models cannot explain the abundance patterns of the
966: metals except for C observed in HMP and EMP stars.
967:
968:
969: \section{SUMMARY}
970:
971: We investigate the evolution of dust formed at Population III SN
972: explosions through the collision with the reverse shocks within SNRs.
973: We adopt the models of dust grains obtained from the calculation of dust
974: formation in Population III SNe by Nozawa et al. (2003) and take into
975: account their spatial distribution and size distribution as the initial
976: condition.
977: The calculations carefully treat the dynamics, erosion, and heating of
978: dust grains, and the evolution of temperature and density of gas in
979: spherically symmetric shocks is solved as a function of time.
980: We also discuss the effect of the hydrogen envelope on the evolution
981: of dust in the SNR.
982: Furthermore, from the analysis of the transport and destruction of dust
983: within SNRs, we investigate the abundances of elements related to dust
984: in the second-generation stars formed in the dense shell of primordial
985: SNRs.
986:
987: The our main results are summarized as follows.
988:
989: 1. The time that the reverse shock encounters the dust that condensed
990: inside the He core depends on the thickness of the H-envelope that was
991: retained by the progenitors of the Population III SNe.
992: If the progenitor stars did not undergo significant mass loss, the
993: reverse shock will encounter the dust $\simeq$$10^3$--10$^4$ yr after
994: the explosion, depending on the density of gas in the ISM.
995:
996: 2. Once dust grains inside the He core collide with the reverse shocks,
997: they will follow different trajectories depending on their initial
998: sizes, resulting in the differential transport and destruction of dust in
999: SNRs.
1000: For Type II SNRs expanding into the ISM with the density of
1001: $n_{\rm H,0}=1$ cm$^{-3}$, small grains with $a_{\rm ini} \la 0.05$
1002: $\micron$ are rapidly trapped into the postshock flow and are
1003: completely destroyed by sputtering.
1004: Grains with $a_{\rm ini}=$ 0.05--0.2 $\micron$ are trapped and remain
1005: in the dense shell behind the forward shock.
1006: Very large grains with $a_{\rm ini} \ga 0.2$ $\micron$ are ejected to
1007: the ISM through the forward shock without significantly decreasing
1008: their sizes.
1009:
1010: 3. The critical size below which dust is completely destroyed in SNRs
1011: is sensitive to the gas density in the ambient medium, and spans the
1012: range of 0.01--0.2 $\micron$ for $n_{\rm H, 0} =$ 0.1--10 cm$^{-3}$.
1013: The resulting size distribution of survived dust is greatly deficient
1014: in small-sized grains, compared with that at the time of dust formation,
1015: although the erosion of large grains produces the smaller-sized grains.
1016: Thus, the mass of dust injected from SNe into the ISM is dominated by
1017: the large grains.
1018:
1019: 4. The total mass fraction of dust destroyed in SNRs ranges from 0.2 to
1020: 1 and increases with increasing the ambient gas density and explosion
1021: energy of SNe.
1022: The destruction efficiency of each dust species is very sensitive to
1023: the initial size distribution, and the dust species whose mass is
1024: predominantly occupied by the sizes larger than the critical size can
1025: survive.
1026: Therefore, the chemical composition, size distribution, and amount of
1027: dust grains supplied from SNe to the ISM are quite different from those
1028: at the time of dust formation.
1029:
1030: 5. The results for the evolution of dust in SNRs presented in this
1031: paper can be directly applied for the initial metallicity of progenitor
1032: stars and ambient medium less than $Z \le 10^{-3}$ $Z_\odot$, and could
1033: be useful for evalulating the evolution of dust in the Galactic SNRs
1034: generated from SNe II.
1035:
1036: 6. The fates of dust grains formed in the ejecta strongly depend on the
1037: thickness of the hydrogen envelope.
1038: For the SNR generated from the SN with the very thin hydrogen envelope,
1039: the collision time of the reverse shock with dust grains inside the He
1040: core is much earlier.
1041: As long as the model of dust for SNe II is employed for the
1042: calculation, the mass of dust supplied to the ISM is larger than SNe II
1043: with the thick hydrogen envelopes.
1044:
1045: 7. If the elemental compositions of dust grains piled up in the SN
1046: shell reflect the metal abundance patterns of the second-generation
1047: stars formed in the dense shell of primordial SNRs, the dust formed in
1048: the unmixed ejecta of SNe II can be responsible for the peculiar
1049: abundance patterns of Mg and Si in HMP stars.
1050: However, another scenario could be necessary to produce the large
1051: overabundances of C and O observed in HMP stars.
1052:
1053:
1054: \acknowledgments
1055:
1056: The authors are grateful to the anonymous referee for critical comments
1057: that are useful for improving the manuscript.
1058: This work has been supported in part by a Grant-in-Aid for Scientific
1059: Research from the Japan Society for the Promotion of Sciences
1060: (16340051 and 18104003).
1061:
1062:
1063: \begin{thebibliography}{}
1064:
1065: \bibitem[Arendt, Dwek \& Moseley(1999)]{are99} Arendt, R. G., Dwek, E.,
1066: \& Moseley, S. H. 1999, \apj, 521, 234
1067: \bibitem[Baraffe et al.(2001)]{bar01} Baraffe, I., Heger, A., \&
1068: Woosley, S. E. 2001. \apj, 550, 890
1069: \bibitem[Beelen et al.(2006)]{bee06} Beelen, A., Cox, P., Benford,
1070: D. J., Dowell, C. D., Kovacs, A.,
1071: Bertoldi, F., Omont, A., \& Carilli, C.,
1072: L. 2006, \apj, 642, 694
1073: \bibitem[Bertoldi et al.(2003)]{ber03} Bertoldi, F., Carilli, C. L.,
1074: Cox, P., Fan, X., Strauss, M. A.,
1075: Beelen, A., Omont, A., \& Zylka, R.
1076: 2003, \aap, 406, L55
1077: \bibitem[Bianchi \& Schneider(2007)]{bia07} Bianchi, S., \& Schneider, R.
1078: 2007, \mnras, in press
1079: (astrp-ph/0704.0586)
1080: \bibitem[Bromm et al.(2001)]{bro01} Bromm, V., Ferrara, A., Coppi, P. S., \&
1081: Larson, R. B. 2001, \mnras, 328, 969
1082: \bibitem[Cazaux \& Spaans(2004)]{caz04} Cazaux, S., \& Spaans, M.
1083: 2004, \apj, 611, 40
1084: \bibitem[Christlieb et al.(2002)]{chr02} Christlieb, N., et al. 2002,
1085: \nat, 419, 904
1086: \bibitem[Christlieb et al.(2004)]{chr04} Christlieb, N., Gustafsson, B.,
1087: Korn, A. J., Barklem, P. S., Beers, T. C.,
1088: Bessell, M. S., Karlsson, T., \&
1089: Mizuno-Wiedner, M. 2004, \apj, 603, 708
1090: \bibitem[Douvion et al.(2001)]{dou01} Douvion, T., Lagage, P. O., \&
1091: Pantin, E. 2001, \aap, 369, 589
1092: \bibitem[Draine \& Salpeter(1979)]{dra79} Draine, B. T., \& Salpeter, E. E.
1093: 1979, \apj, 231, 77
1094: \bibitem[Dwek et al.(1987)]{dwe87} Dwek, E.,, Dinerstein, H. L.,
1095: Gillett, F. C., Hauser, M. G., \& Rice,
1096: W. L. 1987, \apj, 315, 571
1097: \bibitem[Dwek \& Arendt(1992)]{dwe92} Dwek, E., \& Arendt, R. G. 1992,
1098: \araa, 30, 11
1099: \bibitem[Dwek et al. (2007)]{dwe07} Dwek, E., Galliano, F., \& Jones,
1100: A. P. 2007, ApJ, in press
1101: \bibitem[Ennis et al.(2006)]{eni06} Ennis, J. A., Rudnick, L., Reach, W.
1102: T., Smith, J. D., Rho, J., Delaney, T.,
1103: Gomez, H., \& Kozasa, T. 2006, \apj, 652, 376
1104: \bibitem[Frebel et al.(2005)]{fre05} Frebel, A., et al. 2005, \nat, 434, 871
1105: \bibitem[Hines et al.(2004)]{hin04} Hines, D. C., et al.
1106: 2004, \apjs, 154, 290
1107: \bibitem[Hines et al.(2006)]{hin06} Hines, D. C., Krause, O., Rieke,
1108: G. H., Fan, X., Blaylock, M., \&
1109: Neugebauer, G. 2006, \apj, 641, L85
1110: \bibitem[Hirashita \& Ferrara(2002)]{hir02} Hirashita, H., \& Ferrara,
1111: A. 2002, \mnras, 337, 921
1112: \bibitem[Hirashita et al.(2005)]{hir05} Hirashita, H., Nozawa, T.,
1113: Kozasa, T., Ishii, T. T., \&
1114: Takeuchi, T. T.
1115: 2005, \mnras, 357, 1077
1116: \bibitem[Iwamoto et al.(2005)]{iwa05} Iwamoto, N., Umeda, H.,
1117: Tominaga, N., Nomoto, K., \&
1118: Maeda, K. 2005, Science, 309, 451
1119: \bibitem[Kitayama et al.(2004)]{kit04} Kitayama, T., Yoshida, N., Susa,
1120: H., \& Umemura, M.
1121: 2004, \apj, 613, 631
1122: \bibitem[Komiya et al.(2007)]{kom07} Komiya, Y., Suda, T., Minaguchi,
1123: H., Shigeyama, T., Aoki, Y., \&
1124: Fujimoto, M. Y. 2007, \apj, in
1125: press (astro-ph/0610670)
1126: \bibitem[Kozasa et al.(1989)]{koz89} Kozasa, T., Hasegawa, H., \&
1127: Nomoto, K. 1989, \apj, 344, 325
1128: \bibitem[Lagage et al.(1996)]{lag96} Lagage, P. O., Claret, A., Ballet,
1129: J., Boulanger, F., Cesarsky, C. J., Cesarsky, D.,
1130: Fransson, C., \& Pollock, A. 1996, \aap, 315, L273
1131: \bibitem[Machida et al.(2005)]{mac05} Machida, M. N., Tomisaka, K.,
1132: Nakamura, F., \& Fujimoto, M. Y.
1133: 2005, \apj, 622, 39
1134: \bibitem[Maiolino et al.(2004a)]{mai04a} Maiolino, R., Oliva, E.,
1135: Ghinassi, F., Pedani, M.,
1136: Mannucci, F., Mujica, R., \&
1137: Juarez, Y.
1138: 2004a, \aap, 420, 889
1139: \bibitem[Maiolino et al.(2004b)]{mai04b} Maiolino, R., Schneider, R.,
1140: Oliva, E., Bianchi, S.,
1141: Ferrara, A., Mannucci, F.,
1142: Pedani, M., \& Roca Sogorb, M.
1143: 2004b, \nat, 431, 533
1144: \bibitem[Mair et al.(1988)]{mai88} Mair, G., M\"{u}llar, E.,
1145: Hillebrandt, W., \& Arnold, C. N. 1988,
1146: \aap, 199, 114
1147: \bibitem[McKee et al.(1987)]{mck87} McKee, C. F., Hollenbach, D. J.,
1148: Seab, C. G., \& Tielens,
1149: A. G. G. M. 1987, \apj, 318, 674
1150: \bibitem[Morgan \& Edmunds(2003)]{mor03} Morgan, H. L., \& Edmunds, M. G.
1151: 2003, \mnras, 343, 427
1152: \bibitem[Nozawa et al.(2006)]{noz06} Nozawa, T., Kozasa, T., \& Habe, A.
1153: 2006, \apj, 648, 435
1154: \bibitem[Nozawa (2003)]{noz03a} Nozawa, T., 2003, Master Thesis,
1155: Hokkaido University
1156: \bibitem[Nozawa et al.(2003)]{noz03b} Nozawa, T., Kozasa, T., Umeda, H.,
1157: Maeda, K., \& Nomoto, K.
1158: 2003, \apj, 598, 785
1159: \bibitem[Omukai et al.(2005)]{omu05} Omukai, K., Tsuribe, T.,
1160: Schneider, R., \& Ferrara, A. 2005,
1161: \apj, 626, 627
1162: \bibitem[Priddey et al.(2003)]{pri03} Priddey, R. S., Isaak, K. G.,
1163: McMahon, R. G., Robson, E. I.,
1164: \& Pearson, C. P.
1165: 2003, \mnras, 344, L74
1166: \bibitem[Raymond \& Smith(1977)]{ray77} Raymond, J. C., \& Smith, B. W.
1167: 1977, \apjs, 35, 419
1168: \bibitem[Robson et al.(2004)]{rob04} Robson, I., Priddey, R. S., Isaak,
1169: K. G., \& McMahon, R. G.
1170: 2004, \mnras, 351, L29
1171: \bibitem[Schneider et al.(2004)]{sch04} Schneider, R., Ferrara, A., \&
1172: Salvaterra, R.
1173: 2004, \mnras, 351, 1379
1174: \bibitem[Schneider et al.(2006)]{sch06} Schneider, R., Omukai, K.,
1175: Inoue, A. K., \& Ferrara, A. 2006, \mnras,
1176: 369, 1437
1177: \bibitem[Shigeyama et al.(2004)]{shi03} Shigeyama, T., Tsujimoto, T., \&
1178: Yoshii, Y.
1179: 2003, \apj, 586, L57
1180: \bibitem[Smith \& Owocki (2006)]{smi06} Smith, N., \& Owocki,
1181: S. P. 2006, \apj, 645, L45
1182: \bibitem[Smith et al. (2001)]{smi01} Smith, R. K., Brickhouse, N. S.,
1183: Liedahl, D. A., \& Raymond, J. C.
1184: 2001, \apj, 556, L91
1185: \bibitem[Suda et al. (2004)]{sud04} Suda, T., Aikawa, M., Machida,
1186: M. N., Fujimoto, M. Y., \& Iben Jr., I.
1187: 2004, \apj, 611, 476
1188: \bibitem[Sutherland \& Dopita(1993)]{sut93} Sutherland, R. S., \&
1189: Dopita, M. A. 1993, \apjs, 88, 253
1190: \bibitem[Todini \& Ferrara(2001)]{tod01} Todini, P., \& Ferrara, A.
1191: 2001, \mnras, 325, 726
1192: \bibitem[Truelove \& McKee(1999)]{tru99} Truelove, J. K., \& McKee, C. F.
1193: 1999, \apjs, 120, 299
1194: \bibitem[Umeda \& Nomoto(2002)]{ume02} Umeda, H., \& Nomoto, K.
1195: 2002, \apj, 565, 385
1196: \bibitem[Umeda \& Nomoto(2005)]{ume05} Umeda, H., \& Nomoto, K.
1197: 2005, \apj, 619, 427
1198: \bibitem[van Albada et al.(1982)]{van82} van Albada,
1199: G. D., van Leer, B., \& Roberts,
1200: W. W. Jr. 1982, \aap, 108, 76
1201: \bibitem[Venkatesan et al.(2006)]{ven06} Venkatesan, A., Nath, B., B.,
1202: \& Shull, J. M. 2006, \apj, 640, 31
1203: \bibitem[Young et al.(2006)]{ven06} Young, P. A., Fryer, C. L.,
1204: Hungerford, A., Arnett, D.,
1205: Rockfeller, G., Timmes, F. X.,
1206: Voit, B., Meakin, C., \& Eriksen, K.
1207: 2006, \apj, 640, 891
1208: \end{thebibliography}
1209:
1210:
1211: \clearpage
1212:
1213: \begin{figure}
1214: \epsscale{0.7}
1215: \plotone{f1.eps}
1216: \caption{
1217: The structures of (a) the density and (b) the temperature of the gas at
1218: given times within the SNR generated from the explsoion of star with
1219: $M_{\rm pr} = 20$ $M_\odot$ and expanding into the ISM with
1220: $n_{\rm H,0}=1$ cm$^{-3}$ (the standard model).
1221: The positions of the forward and reverse shocks are indicated by the
1222: downward arrows in (a) and (b), respectively.
1223: \textit{See the electronic edition of the Journal for a color version
1224: of this figure.}\label{fig1}}
1225: \end{figure}
1226:
1227:
1228: \clearpage
1229:
1230: \begin{figure}
1231: \epsscale{0.7}
1232: \plotone{f2.eps}
1233: \caption{
1234: The time evolution of (a) the position $R_{\rm rs}$ and (b) the
1235: velocity $V_{\rm rs}$ of the reverse shock.
1236: The thick solid curves represents the results for the standard model
1237: with $M_{\rm pr} = 20$ $M_\odot$ and $n_{\rm H,0}=1$ cm$^{-3}$, and the
1238: thin solid curves for the model with $M_{\rm pr} = 170$ $M_\odot$ and
1239: $n_{\rm H,0}=1$ cm$^{-3}$.
1240: The results that do not include the cooling by thermal emission from
1241: dust are also shown by the thick and thin dashed curves for
1242: $M_{\rm pr} = 20$ $M_\odot$ and 170 $M_\odot$,
1243: respectively.
1244: \textit{See the electronic edition of the Journal for a color version
1245: of this figure.}\label{fig2}}
1246: \end{figure}
1247:
1248:
1249: \clearpage
1250:
1251: \begin{figure}
1252: \epsscale{0.7}
1253: \plotone{f3.eps}
1254: \caption{
1255: The time evolutions of (a) the positions and (b) the ratios of size to
1256: the initial size of dust grains within the SNR for the standard model.
1257: The positions of the forward and reverse shocks are depicted by the
1258: thick solid curves in (a), along with the position of the surface of
1259: the He core.
1260: Among nine dust species for the unmixed grain model, are shown C,
1261: Mg$_2$SiO$_4$, and Fe grains with the initial sizes of 0.01 $\micron$
1262: (\textit{dotted lines}), 0.1 $\micron$ (\textit{solid lines}), and 1
1263: $\micron$ (\textit{dashed lines}), respectively.
1264: \textit{See the electronic edition of the Journal for a color version
1265: of this figure.}\label{fig2}}
1266: \end{figure}
1267:
1268:
1269: \clearpage
1270:
1271: \begin{figure}
1272: \epsscale{0.7}
1273: \plotone{f4.eps}
1274: \caption{
1275: The collision time $t_{\rm coll}$ of the reverse shocks with the He
1276: core vs. the progenitor stellar mass for different gas density in the
1277: ISM.
1278: The results for $n_{\rm H,0}=$ 0.1, 1, and 10 cm$^{-3}$ are connected
1279: by the dot-dashed, dashed, and dotted lines, respectively.
1280: \textit{See the electronic edition of the Journal for a color version
1281: of this figure.}\label{fig4}}
1282: \end{figure}
1283:
1284:
1285: \clearpage
1286:
1287: \begin{figure}
1288: \epsscale{0.7}
1289: \plotone{f5.eps}
1290: \caption{
1291: The ratio of the final size $a_{\rm fin}$ to the initial size
1292: $a_{\rm ini}$ of dust eroded by the kinetic sputtering in the ISM
1293: with the primordial gas composition as a function of the escape
1294: velocity $w_0$, where the final relative velocity $w_{\rm fin}$ is
1295: taken as 10 km s$^{-1}$.
1296: The results are shown for C, Si, Mg$_2$SiO$_4$, Fe$_3$O$_4$, and Fe
1297: grains.
1298: \textit{See the electronic edition of the Journal for a color version
1299: of this figure.}\label{fig7}}
1300: \end{figure}
1301:
1302:
1303: \clearpage
1304:
1305: \begin{figure}
1306: \epsscale{0.7}
1307: \plotone{f6a.eps}
1308: \plotone{f6b.eps}
1309: \caption{
1310: The size distribution of each dust species for the standard model;
1311: (a) for the initial size distribution before destruction and
1312: (b) for the resulting size distribution of survived dust after
1313: destruction.
1314: \textit{See the electronic edition of the Journal for a color version
1315: of this figure.}\label{fig6}}
1316: \end{figure}
1317:
1318:
1319: \clearpage
1320:
1321: \begin{figure}
1322: \epsscale{0.7}
1323: \plotone{f7a.eps}
1324: \plotone{f7b.eps}
1325: \caption{
1326: The total mass of survived dust for the various progenitor mass and the
1327: gas density in the ISM (a) for the unmixed grain model and (b) for the
1328: mixed grain model.
1329: The results for $n_{\rm H,0}=$ 0.1, 1, and 10 cm$^{-3}$ are connected
1330: by the dot-dashed, dashed, and dotted lines, respectively.
1331: The solid lines are for the initial total mass of dust at the time of
1332: dust formation.
1333: \textit{See the electronic edition of the Journal for a color version
1334: of this figure.}\label{fig10}}
1335: \end{figure}
1336:
1337:
1338: \clearpage
1339:
1340: \begin{figure}
1341: \epsscale{0.7}
1342: \plotone{f8.eps}
1343: \caption{
1344: The time evolutions of (a) the positions and (b) the ratios of size to
1345: the initial size of dust grains within the SNR from the SN with the
1346: hydrogen envelope of 0.7 $M_\odot$ and the explosion energy of
1347: 10$^{51}$ erg for $n_{\rm H,0} = 1$ cm$^{-3}$.
1348: The positions of the forward and reverse shocks are depicted by the
1349: thick solid curves in (a), along with the position of the surface of
1350: the He core.
1351: Among nine dust species for the unmixed grain model, are shown C,
1352: Mg$_2$SiO$_4$, and Fe grains with the initial sizes of 0.01 $\micron$
1353: (\textit{dotted lines}) and 0.1 $\micron$ (\textit{solid lines}),
1354: respectively.
1355: \textit{See the electronic edition of the Journal for a color version
1356: of this figure.}\label{fig7}}
1357: \end{figure}
1358:
1359:
1360: \clearpage
1361:
1362: \begin{deluxetable}{lccccc}
1363: \tablewidth{0pt}
1364: \tablecaption{THE MAIN RESULTS OF CALCULATIONS FOR THE UNMIXED GRAIN MODEL}
1365: \tablehead{
1366: \colhead{$M_{\rm pr}$} &
1367: \colhead{} & \colhead{} & \colhead{} &
1368: \colhead{$t_{\rm tr}$} & \colhead{$M_{\rm shell}^{\rm H}$} \\
1369: \colhead{($M_{\odot}$)} &
1370: \colhead{$\epsilon_{\rm dest}$} & \colhead{$\epsilon_{\rm eject}$} &
1371: \colhead{$\epsilon_{\rm shell}$} &
1372: \colhead{($10^6$ yr)} & \colhead{($10^4$ $M_{\odot}$)}
1373: }
1374: \startdata
1375: \multicolumn{6}{c}{$n_{\rm H,0}=0.1$ cm$^{-3}$} \\
1376: \tableline
1377: 13 & 0.240 & 0.756 & 0.004 & 2.13 & 2.96 \\
1378: 20 & 0.386 & 0.610 & 0.004 & 2.05 & 2.58 \\
1379: 25 & 0.198 & 0.800 & 0.002 & 2.12 & 2.89 \\
1380: 30 & 0.383 & 0.614 & 0.003 & 2.11 & 2.83 \\
1381: 170 & 0.562 & 0.438 & $<$0.001 & 4.58 & 34.4 \\
1382: 200 & 0.746 & 0.247 & 0.007 & 5.13 & 50.2 \\
1383: \tableline
1384: \tableline
1385: \multicolumn{6}{c}{$n_{\rm H,0}=1$ cm$^{-3}$} \\
1386: \tableline
1387: 13 & 0.698 & 0.256 & 0.046 & 0.814 & 2.07 \\
1388: 20 & 0.752 & 0.231 & 0.017 & 0.767 & 1.72 \\
1389: 25 & 0.517 & 0.474 & 0.009 & 0.800 & 1.97 \\
1390: 30 & 0.785 & 0.206 & 0.009 & 0.790 & 1.88 \\
1391: 170 & 0.958 & 0.040 & 0.002 & 1.81 & 25.4 \\
1392: 200 & 0.997 & 0.002 & 0.001 & 2.02 & 35.2 \\
1393: \tableline
1394: \tableline
1395: \multicolumn{6}{c}{$n_{\rm H,0}=10$ cm$^{-3}$} \\
1396: \tableline
1397: 13 & 0.936 & 0.004 & 0.060 & 0.308 & 1.31 \\
1398: 20 & 0.942 & 0.013 & 0.045 & 0.301 & 1.21 \\
1399: 25 & 0.851 & 0.103 & 0.046 & 0.310 & 1.33 \\
1400: 30 & 0.920 & 0.042 & 0.038 & 0.309 & 1.31 \\
1401: 170 & 0.999 & $<$0.001 & $<$0.001 & 0.723 & 18.0 \\
1402: 200 & 1.00 & $<$0.001 & $<$0.001 & 0.825 & 27.1 \\
1403: \enddata
1404: \tablecomments{
1405: For a given $n_{\rm H,0}$, the mass fraction of dust destroyed
1406: $\epsilon_{\rm dest}$, piled up in the dense shell
1407: $\epsilon_{\rm shell}$, and ejected to the ISM $\epsilon_{\rm eject}$
1408: as a function of the progenitor mass $M_{\rm pr}$ for the unmixed grain
1409: model.
1410: The truncation time $t_{\rm tr}$ and the hydrogen mass $M^{\rm H}_{\rm
1411: shell}$ in the dense SN shell are given by the units of $10^6$
1412: yr and $10^4$ $M_\odot$, respectively
1413: }
1414: \end{deluxetable}
1415:
1416:
1417: \clearpage
1418:
1419: \begin{deluxetable}{lccccc}
1420: \tablewidth{0pt}
1421: \tablecaption{THE MAIN RESULTS OF CALCULATIONS FOR THE MIXED GRAIN MODEL}
1422: \tablehead{
1423: \colhead{$M_{\rm pr}$} &
1424: \colhead{} & \colhead{} & \colhead{} &
1425: \colhead{$t_{\rm tr}$} & \colhead{$M_{\rm shell}^{\rm H}$} \\
1426: \colhead{($M_{\odot}$)} &
1427: \colhead{$\epsilon_{\rm dest}$} & \colhead{$\epsilon_{\rm eject}$} &
1428: \colhead{$\epsilon_{\rm shell}$} &
1429: \colhead{($10^6$ yr)} & \colhead{($10^4$ $ M_{\odot}$)}
1430: }
1431: \startdata
1432: \multicolumn{6}{c}{$n_{\rm H,0}=0.1$ cm$^{-3}$} \\
1433: \tableline
1434: 13 & 0.775 & 0.212 & 0.013 & 2.14 & 2.96 \\
1435: 20 & 0.664 & 0.311 & 0.025 & 2.05 & 2.57 \\
1436: 25 & 0.747 & 0.245 & 0.008 & 2.10 & 2.82 \\
1437: 30 & 0.569 & 0.415 & 0.016 & 2.10 & 2.80 \\
1438: 170 & 0.742 & 0.258 & $<$0.001 & 4.55 & 33.7 \\
1439: 200 & 0.751 & 0.248 & 0.001 & 5.12 & 49.8 \\
1440: \tableline
1441: \tableline
1442: \multicolumn{6}{c}{$n_{\rm H,0}=1$ cm$^{-3}$} \\
1443: \tableline
1444: 13 & 0.995 & 0.001 & 0.004 & 0.814 & 2.07 \\
1445: 20 & 0.907 & 0.075 & 0.018 & 0.767 & 1.71 \\
1446: 25 & 0.988 & 0.009 & 0.003 & 0.800 & 1.95 \\
1447: 30 & 0.920 & 0.060 & 0.020 & 0.788 & 1.86 \\
1448: 170 & 0.993 & 0.007 & $<$0.001 & 1.81 & 24.8 \\
1449: 200 & 0.989 & 0.011 & $<$0.001 & 2.00 & 34.2 \\
1450: \tableline
1451: \tableline
1452: \multicolumn{6}{c}{$n_{\rm H,0}=10$ cm$^{-3}$} \\
1453: \tableline
1454: 13 & 1.00 & 0. & $<$0.001 & 0.309 & 1.33 \\
1455: 20 & 0.988 & 0.002 & 0.010 & 0.301 & 1.21 \\
1456: 25 & 1.00 & 0. & $<$0.001 & 0.316 & 1.38 \\
1457: 30 & 0.995 & $<$0.001 & 0.005 & 0.308 & 1.29 \\
1458: 170 & 1.00 & 0. & $<$0.001 & 0.726 & 17.8 \\
1459: 200 & 1.00 & 0. & 0. & 0.817 & 26.0 \\
1460: \enddata
1461: \tablecomments{
1462: Same as Table 1, but for the mixed grain model.
1463: }
1464: \end{deluxetable}
1465:
1466:
1467: \clearpage
1468:
1469: \begin{deluxetable}{lccccccccc}
1470: \tablewidth{0pt}
1471: \tablecaption{THE DESTRUCTION EFFICIENCY OF EACH DUST SPECIES FOR THE
1472: UNMIXED GRAIN MODEL}
1473: \tablehead{
1474: \colhead{$M_{\rm pr}$} &
1475: \colhead{} & \colhead{} & \colhead{} & \colhead{} & \colhead{} &
1476: \colhead{} & \colhead{} & \colhead{} & \colhead{} \\
1477: \colhead{($M_{\odot}$)} &
1478: \colhead{Fe} & \colhead{Si} & \colhead{FeS} & \colhead{MgSiO$_3$} &
1479: \colhead{SiO$_2$} & \colhead{Al$_2$O$_3$} &
1480: \colhead{Mg$_2$SiO$_4$} & \colhead{MgO} & \colhead{C}
1481: }
1482: \startdata
1483: \multicolumn{10}{c}{$n_{\rm H,0}=0.1$ cm$^{-3}$} \\
1484: \tableline
1485: 13 & 0.209 & 0.257 & 0.939 & 0.892 & 0.297
1486: & 1.00 & 0.349 & 0.615 & 0.048 \\
1487: 20 & 0.210 & 0.127 & 0.881 & 0.945 & 0.518
1488: & 1.00 & 0.437 & 0.440 & 0.146 \\
1489: 25 & 0.089 & 0.061 & 0.693 & 0.570 & 0.086
1490: & 0.999 & 0.201 & 0.520 & 0.080 \\
1491: 30 & 0.099 & 0.077 & 0.636 & 0.901 & 0.277
1492: & 0.992 & 0.543 & 0.412 & 0.299 \\
1493: 170 & 0.644 & 0.273 & 1.00 & 0.987 & 0.441
1494: & 1.00 & 0.776 & 0.772 & 0.284 \\
1495: 200 & 0.712 & 0.787 & 1.00 & 0.990 & 0.438
1496: & 1.00 & 0.865 & 0.820 & 0.210 \\
1497: \tableline
1498: \tableline
1499: \multicolumn{10}{c}{$n_{\rm H,0}=1$ cm$^{-3}$} \\
1500: \tableline
1501: 13 & 0.715 & 0.660 & 1.00 & 0.998 & 0.882
1502: & 1.00 & 0.964 & 0.997 & 0.520 \\
1503: 20 & 0.661 & 0.493 & 0.999 & 0.999 & 0.926
1504: & 1.00 & 0.961 & 0.925 & 0.447 \\
1505: 25 & 0.379 & 0.289 & 0.984 & 0.968 & 0.369
1506: & 1.00 & 0.719 & 0.934 & 0.284 \\
1507: 30 & 0.405 & 0.349 & 0.970 & 0.997 & 0.745
1508: & 1.00 & 0.989 & 0.946 & 0.842 \\
1509: 170 & 0.999 & 0.872 & 1.00 & 1.00 & 0.998
1510: & 1.00 & 1.00 & 1.00 & 0.985 \\
1511: 200 & 0.989 & 0.999 & 1.00 & 1.00 & 0.992
1512: & 1.00 & 1.00 & 1.00 & 0.996 \\
1513: \tableline
1514: \tableline
1515: \multicolumn{10}{c}{$n_{\rm H,0}=10$ cm$^{-3}$} \\
1516: \tableline
1517: 13 & 0.958 & 0.894 & 1.00 & 1.00 & 0.995
1518: & 1.00 & 1.00 & 1.00 & 0.893 \\
1519: 20 & 0.961 & 0.851 & 1.00 & 1.00 & 1.00
1520: & 1.00 & 1.00 & 1.00 & 0.896 \\
1521: 25 & 0.861 & 0.699 & 1.00 & 1.00 & 1.00
1522: & 1.00 & 0.998 & 1.00 & 0.727 \\
1523: 30 & 0.824 & 0.706 & 1.00 & 1.00 & 0.958
1524: & 1.00 & 1.00 & 1.00 & 0.990 \\
1525: 170 & 1.00 & 0.999 & 1.00 & 1.00 & 1.00
1526: & 1.00 & 1.00 & 1.00 & 1.00 \\
1527: P200 & 1.00 & 1.00 & 1.00 & 1.00 & 1.00
1528: & 1.00 & 1.00 & 1.00 & 1.00 \\
1529: \enddata
1530: \tablecomments{
1531: For a given $n_{\rm H,0}$, the mass fraction of dust destroyed
1532: as a function of the progenitor mass $M_{\rm pr}$ for each grain
1533: species in the unmixed grain model.
1534: }
1535: \end{deluxetable}
1536:
1537:
1538: \clearpage
1539:
1540: \begin{deluxetable}{lccccc}
1541: \tablewidth{0pt}
1542: \tablecaption{THE DESTRUCTION EFFICIENCY OF EACH DUST SPECIES FOR THE
1543: MIXED GRAIN MODEL}
1544: \tablehead{
1545: \colhead{$M_{\rm pr}$} &
1546: \colhead{} & \colhead{} &
1547: \colhead{} & \colhead{} & \colhead{} \\
1548: \colhead{($M_{\odot}$)} &
1549: \colhead{Al$_2$O$_3$} & \colhead{MgSiO$_3$} &
1550: \colhead{Mg$_2$SiO$_4$} & \colhead{SiO$_2$} & \colhead{Fe$_3$O$_4$}
1551: }
1552: \startdata
1553: \multicolumn{6}{c}{$n_{\rm H,0}=0.1$ cm$^{-3}$} \\
1554: \tableline
1555: 13 & 1.00 & 0.862 & 0.868 & 0.676 & 0.788 \\
1556: 20 & 1.00 & 0.913 & 0.788 & 0.518 & 0.992 \\
1557: 25 & 1.00 & 0.910 & 0.877 & 0.596 & 0.999 \\
1558: 30 & 0.997 & 0.802 & 0.620 & 0.461 & 0.999 \\
1559: 170 & 1.00 & 0.961 & 0.957 & 0.585 & 1.00 \\
1560: 200 & 1.00 & 0.984 & 0.995 & 0.615 & 0.999 \\
1561: \tableline
1562: \tableline
1563: \multicolumn{6}{c}{$n_{\rm H,0}=1$ cm$^{-3}$} \\
1564: \tableline
1565: 13 & 1.00 & 0.999 & 0.999 & 0.988 & 0.999 \\
1566: 20 & 1.00 & 0.996 & 0.977 & 0.850 & 1.00 \\
1567: 25 & 1.00 & 1.00 & 1.00 & 0.976 & 1.00 \\
1568: 30 & 1.00 & 0.994 & 0.951 & 0.875 & 1.00 \\
1569: 170 & 1.00 & 1.00 & 1.00 & 0.989 & 1.00 \\
1570: 200 & 1.00 & 1.00 & 1.00 & 0.983 & 1.00 \\
1571: \tableline
1572: \tableline
1573: \multicolumn{6}{c}{$n_{\rm H,0}=10$ cm$^{-3}$} \\
1574: \tableline
1575: 13 & 1.00 & 1.00 & 1.00 & 0.999 & 1.00 \\
1576: 20 & 1.00 & 1.00 & 1.00 & 0.980 & 1.00 \\
1577: 25 & 1.00 & 1.00 & 1.00 & 1.00 & 1.00 \\
1578: 30 & 1.00 & 1.00 & 0.996 & 0.976 & 1.00 \\
1579: 170 & 1.00 & 1.00 & 1.00 & 1.00 & 1.00 \\
1580: P200 & 1.00 & 1.00 & 1.00 & 1.00 & 1.00 \\
1581: \enddata
1582: \tablecomments{
1583: Same as Table 3, but for each grain species in the mixed grain model.
1584: }
1585: \end{deluxetable}
1586:
1587:
1588: \clearpage
1589:
1590: \begin{deluxetable}{lccccc}
1591: \tablewidth{0pt}
1592: \tablecaption{ELEMENTAL ABUNDANCES IN SN SHELL}
1593: \tablehead{
1594: \colhead{$M_{\rm pr}$} &
1595: \colhead{} & \colhead{} &
1596: \colhead{} & \colhead{} & \colhead{} \\
1597: \colhead{($M_{\odot}$)} &
1598: \colhead{[Fe/H]} & \colhead{[C/Fe]} &
1599: \colhead{[O/Fe]} & \colhead{[Mg/Fe]} &
1600: \colhead{[Si/Fe]}
1601: }
1602: \startdata
1603: \multicolumn{6}{c}{$n_{\rm H,0}=0.1$ cm$^{-3}$} \\
1604: \tableline
1605: 13 & $-6.43$ & $-0.274$ & $-0.699$ & $-0.230$ & $1.92$ \\
1606: 20 & $-5.20$ & $0.117$ & $-0.595$ & $0.034$ & $0.410$ \\
1607: 25 & $-5.90$ & $1.11$ & $-1.42$ & $-0.500$ & $-0.552$ \\
1608: 30 & $-5.56$ & $0.566$ & $-0.043$ & $0.739$ & $0.866$ \\
1609: \tableline
1610: \tableline
1611: \multicolumn{6}{c}{$n_{\rm H,0}=1$ cm$^{-3}$} \\
1612: \tableline
1613: 13 & $-5.15$ & $1.11$ & $-0.555$ & $-0.459$ & $1.01$ \\
1614: 20 & $-5.53$ & $0.992$ & $0.585$ & $1.16$ & $1.87$ \\
1615: 25 & $-5.23$ & $1.09$ & $-0.412$ & $0.407$ & $0.989$ \\
1616: 30 & $-5.11$ & $0.797$ & $0.242$ & $1.09$ & $1.26$ \\
1617: \tableline
1618: \tableline
1619: \multicolumn{6}{c}{$n_{\rm H,0}=10$ cm$^{-3}$} \\
1620: \tableline
1621: 13 & $-4.13$ & $0.284$ & $-2.54$ & $-3.89$ & $0.599$ \\
1622: 20 & $-4.92$ & $0.946$ & $-2.15$ & $-1.80$ & $2.14$ \\
1623: 25 & $-5.10$ & $1.60$ & $0.122$ & $0.232$ & $2.34$ \\
1624: 30 & $-5.11$ & $-0.207$ & $0.375$ & $-1.23$ & $2.66$ \\
1625: \enddata
1626: \tablecomments{
1627: [Fe/H] and abundances of C, O, Mg, and Si relative to Fe in the shell
1628: of primordial Type II SNRs for given $M_{\rm pr}$ and $n_{\rm H,0}$
1629: with the unmixed grain model.
1630: }
1631: \end{deluxetable}
1632:
1633:
1634: %\clearpage
1635: %
1636: %\begin{deluxetable}{lcccc}
1637: %\tablewidth{0pt}
1638: %\tablecaption{ELEMENTAL ABUNDANCE IN SN SHELL FOR THE MIXED CASE}
1639: %\tablehead{
1640: %\colhead{$M_{\rm pr}$} &
1641: %\colhead{} & \colhead{} &
1642: %\colhead{} & \colhead{} \\
1643: %\colhead{($M_{\odot}$)} &
1644: %\colhead{[Fe/H]} & \colhead{[O/Fe]} &
1645: %\colhead{[Mg/Fe]} & \colhead{[Si/Fe]}
1646: %}
1647: %\startdata
1648: %\multicolumn{5}{c}{$n_{\rm H,0}=0.1$ cm$^{-3}$} \\
1649: %\tableline
1650: %13 & $-5.74$ & $0.253$ & $0.782$ & $1.24$ \\
1651: %20 & $-5.20$ & $0.637$ & $1.11$ & $1.64$ \\
1652: %25 & $-7.53$ & $2.55$ & $3.03$ & $3.56$ \\
1653: %30 & $-6.31$ & $1.83$ & $2.61$ & $2.77$ \\
1654: %\enddata
1655: %\tablecomments{
1656: % For $n_{\rm H,0} =$ 0.1 cm$^{-3}$, the elemental abundances of Fe, O,
1657: % Mg, and Si in the shell of primordial Type II SNRs as a function of the
1658: % progenitor mass $M_{\rm pr}$ for the mixed grain model.
1659: %}
1660: %\end{deluxetable}
1661:
1662: \end{document}
1663:
1664:
1665: