1: %\documentclass[preprint]{aastex}
2: %\documentclass[12pt,preprint]{aastex} %a one-column, single-spaced document
3: %\documentclass[manuscript]{aastex} %a one-column, double-spaced document
4: \documentclass[preprint2]{aastex} %a double-column, single-spaced document:
5:
6: %\slugcomment{Submitted to the Astrophysical Journal}
7:
8: \shorttitle{\textit{r}-Process in Anisotropic Neutrino Winds}
9:
10: \shortauthors{Wanajo}
11:
12: \begin{document}
13:
14: \title{The {\boldmath \textit{r}}-Process in the Proto-Neutron-Star
15: Winds with Anisotropic Neutrino Emission}
16:
17: \author{Shinya Wanajo}
18:
19: \affil{Department of Astronomy, School of Science,
20: University of Tokyo, Bunkyo-ku, Tokyo, 113-8654, Japan;
21: wanajo@astron.s.u-tokyo.ac.jp}
22:
23: \begin{abstract}
24: The astrophysical origin of the \textit{r}-process nuclei is still
25: unknown. Even the most promising scenario, the neutrino-driven winds
26: from a nascent neutron star, encounters severe difficulties in
27: obtaining requisite entropy and short dynamic timescale for the
28: \textit{r}-process. In this study, the effect of anisotropy in
29: neutrino emission from a proto-neutron star surface is examined with
30: semi-analytic neutrino-driven wind models. The increase of neutrino
31: number density in the wind owing to the anisotropy is modeled
32: schematically by enhancing the \textit{effective} neutrino
33: luminosity. It is shown that the neutrino heating rate from
34: neutrino-antineutrino pair annihilation into electron-positron pairs
35: can significantly increase owing to the anisotropy and play a dominant
36: role for the heating of wind material. A factor of five increase in
37: the effective neutrino luminosity results in $50 \%$ higher entropy
38: and a factor of ten shorter dynamic timescale owing to this enhanced
39: neutrino heating. The nucleosynthesis calculations show that this
40: change is enough for the robust \textit{r}-process, producing the
41: third abundance peak ($A = 195$) and beyond. Future multi-dimensional
42: studies with accurate neutrino transport will be needed if such
43: anisotropy relevant for the current scenario (more than a factor of a
44: few) is realized during the wind phase ($\sim 1-10\, \textrm{s}$).
45: \end{abstract}
46:
47: \keywords{
48: nuclear reactions, nucleosynthesis, abundances
49: --- stars: abundances
50: --- stars: neutron
51: --- supernovae: general
52: }
53:
54: \section{Introduction}
55: The astrophysical site of the rapid-neutron-capture nucleosynthesis
56: (\textit{r}-process), which accounts for about half of nuclei heavier
57: than iron, has been a long-standing mystery. During the last decade,
58: the neutrino-heated ejecta from a nascent neutron star
59: \citep[neutrino-driven winds,][]{Woos94} has been considered to be the
60: most promising astrophysical site for the \textit{r}-process. Previous
61: studies show, however, severe problems in obtaining requisite high
62: entropy and short dynamic timescale for the production of heavy
63: \textit{r}-process nuclei \citep{Qian96, Otsu00, Sumi00, Wana01,
64: Thom01}. The general relativistic effect for a very compact
65: proto-neutron star \citep[e.g., the mass of $2.0\, M_\odot$ with the
66: radius of 10~km,][]{Otsu00, Wana01} or a magnetar-like magnetic field
67: strength \citep{Thom03, Suzu05} have been invoked to increase entropy
68: and reduce the dynamic timescale of the winds. It is questionable,
69: however, if such physical conditions can be the general requirements
70: for the \textit{r}-process nucleosynthesis.
71:
72: \citet{Qian96} have suggested that an additional energy input to the
73: neutrino-driven wind at between 1.5 and 3 times the neutron star
74: radius is efficient to increase entropy and reduce dynamic timescale
75: of the wind material. In this \textit{Letter}, it is shown that strong
76: anisotropy in neutrino emission from the proto-neutron star,
77: \textit{if it exists}, acts as this extra energy source and helps the
78: \textit{r}-process. The neutrino-driven wind model with spherically
79: symmetric, steady outflow approximation is used to obtain the wind
80: trajectories (\S~2). A sudden increase of neutrino number density in
81: winds owing to anisotropic neutrino emission is modeled by enhancing
82: the neutrino luminosity. Nucleosynthesis calculations with the
83: obtained thermodynamic trajectories are performed to demonstrate this
84: effect (\S~3). Finally, a possible origin of this anisotropy in
85: neutrino emission and some implications of this study are discussed
86: (\S~4).
87:
88: \section{Wind Models with Anisotropic Neutrino Emission}
89:
90: The wind trajectories in this study are obtained using the
91: semi-analytic, general relativistic model of neutrino-driven winds
92: explored in \citet{Otsu00} and \citet{Wana01, Wana02}. In this model,
93: the system is treated as time stationary and spherically symmetric,
94: and the radius of the neutron star is assumed to be the same as that
95: of the neutrino sphere. The heating source that drives matter from the
96: neutrino sphere is due to neutrino interactions. Heating is due to
97: $\nu_e$ and $\bar{\nu}_e$ capture on free nucleons ($\dot q_{\nu N}$),
98: neutrino scattering by electrons and positrons ($\dot q_{\nu e}$), and
99: neutrino-antineutrino pair annihilation into electron-positron pairs
100: ($\dot q_{\nu \nu}$). Cooling is due to electron and positron capture
101: on free nucleons ($\dot q_{e N}$) and electron-positron pair
102: annihilation into neutrino-antineutrino pairs ($\dot q_{e e}$). The
103: rms average neutrino energies are taken to be 10, 20, and 30~MeV, for
104: electron, anti-electron, and the other flavors of neutrinos,
105: respectively. The mass ejection rate at the neutrino sphere $\dot M$
106: is determined so that the wind becomes supersonic through the sonic
107: point.
108:
109: The mass and radius of the neutron star are taken to be $M = 1.4\,
110: M_\odot$ and $R = 10\, \mathrm{km}$, respectively. The neutrino
111: luminosity of one specific flavor is assumed to be the same for all
112: other flavors, which is taken to be a constant value $L_\nu = 1 \times
113: 10^{51}\, \mathrm{ergs}\, \mathrm{s}^{-1}$. Note that the assumption
114: of a constant $L_\nu$ is reasonable, since the crossing time of a wind
115: over the heating region ($< 30\, \mathrm{km}$, see Fig.~2) is short
116: enough, $\sim 0.1\, \mathrm{s}$, compared to the decay timescale of
117: $L_\nu$ \citep[a few seconds, e.g.,][]{Woos94}. As explored in
118: previous studies, this \textit{typical} choice of parameter set (with
119: isotropic neutrino emission) results in insufficient physical
120: conditions (i.e., low entropy and long dynamic timescale) for the
121: production of heavy \textit{r}-process nuclei \citep[e.g.,][]{Wana01}.
122:
123: In this study, anisotropy in neutrino emission is modeled
124: schematically as follows. Given there is substantially higher neutrino
125: emission from the ``hot spot'', which is marked by the point P$_1$
126: ($\mathrm{OP}_1 = R$) in Figure~1. At the point P$_0$ ($\mathrm{OP}_0
127: = R$) nearby P$_1$, the ejection of matter is due to the local (lower)
128: isotropic neutrino emission around P$_0$. The matter suddenly sees a
129: substantially larger number of neutrinos when passing through the
130: point P$_2$. Note that neutrino emission at an arbitrary point on the
131: neutrino sphere (e.g., P$_0$ or P$_1$) is assumed to be isotropic in
132: all directions (i.e., the local neutrino flux is \textit{not} radial)
133: as in \citet{Otsu00}. This sudden increase of the neutrino number
134: density at P$_2$ is approximated by a jump of the neutrino luminosity
135: from the original value $L_\nu = 1 \times 10^{51}\, \mathrm{ergs}\,
136: \mathrm{s}^{-1}$ for $R < r < R_2$ to the \textit{effective}
137: luminosity $L_{\nu 2}$ for $r \ge R_2$, where $r$ is the distance from
138: the center O and $R_2 = \mathrm{OP}_2$.
139:
140: The wind models considered in this study are listed in the first
141: column of Table~1, where $R_2$ (second column) and $L_{\nu 2}$ (third
142: column) are taken to be 10, 12, 15, and 20~km, and 1.0, 2.0, 3.0, 4.0,
143: and 5.0 in units of $10^{51}\, \mathrm{ergs}\, \mathrm{s}^{-1}$. The
144: resulting net heating rate ($\dot q$; \textit{top panel}) for models
145: A1, A5, B5, C5, and D5 and each heating/cooling rate (\textit{bottom
146: panel}) for models A5 and B5 are shown in Figure~2, as functions of
147: $r$. Note that A1-A5 are \textit{isotropic} wind models (i.e., $R_2 =
148: R$).
149:
150: For isotropic winds (A1-A5 in Table~1), the higher $L_{\nu 2}$ ($=
151: L_\nu$ in these cases) results in shorter dynamic timescale
152: $\tau_\mathrm{dyn}$ ($\equiv |\rho/(d\rho/dt)|_{T = 0.5\,
153: \mathrm{MeV}}$) and \textit{lower} asymptotic (i.e., maximum) entropy
154: $s$. This shows that the increased $\dot q$ (see A1 and A5 in Fig.~2,
155: \textit{top panel}) is consumed to drive more matter (i.e., higher
156: $\dot M$ as can be seen in Table~1) from the neutron star surface with
157: faster velocity, rather than to increase entropy. In contrast, for
158: anisotropic models, an increase of $L_{\nu 2}$ (for $r \ge R_2$) is
159: quite efficient \textit{both} to increase entropy and to reduce
160: dynamic timescale (Table~1). The reason is that the matter has been
161: already lifted with low $L_\nu\, (= 1 \times 10^{51}\, \mathrm{ergs}\,
162: \mathrm{s}^{-1} < L_{\nu 2})$ and thus with small $\dot M$. Therefore,
163: the density (and temperature) at arbitrary $r$ is significantly small
164: compared to the corresponding isotropic wind. This can be seen in the
165: 5th (and 6th) column in Table~1, which lists the density $\rho_{13}$
166: (and temperature $T_{13}$) at $r = 13\, \mathrm{km}$ (see about one
167: order difference in $\rho_{13}$ for A5 and B5).
168:
169: For isotropic wind models, the five times greater neutrino luminosity
170: simply results in the increase of $\dot q$ with the same factor (A1
171: and A5 in Fig.~2, \textit{top panel}). This does not hold, however,
172: for anisotropic wind models. For model B5, the maximum $\dot q$ is as
173: twice large as that for model A5 (with the same $L_{\nu 2}$), and more
174: than 10 times larger than that for model A1 (with the same
175: $L_\nu$). This can be explained as follows. As shown in Figure~2
176: (\textit{bottom panel}), for isotropic winds (\textit{dashed lines};
177: A5), the heating is mainly due to $\dot q_{\nu N}$ and $\dot q_{\nu
178: e}$, while $\dot q_{\nu \nu}$ plays only a minor role. In contrast,
179: for anisotropic winds (\textit{solid lines}; B5), the neutrino pair
180: annihilation $\dot q_{\nu \nu}$ plays a crucial role, whose peak (at
181: $r \approx 13\, \mathrm{km}$) is a factor of seven higher than that in
182: A5. This effect can be clearly seen in Figure~2 (\textit{top panel}),
183: in which the case without an increase of $\dot q_{\nu \nu}$
184: (model~B5a) and with an increase of $\dot q_{\nu \nu}$ only
185: (model~B5b) are compared (see also Table~1).
186:
187: This is due to the difference of $\rho$ and $T$ dependences in these
188: heating terms. For a fixed set of $r$, $Y_e$, $L_\nu$ and neutrino
189: mean energies, these heating rates are related to $\rho$ and $T$ such
190: as $\dot q_{\nu N} = \textrm{constant}$, $\dot q_{\nu e} \propto T^4
191: \rho^{-1}$, and $\dot q_{\nu \nu} \propto \rho^{-1}$ \citep{Qian96,
192: Otsu00}. As a result, $\dot q_{\nu N}$ in B5 (Fig.~2, \textit{bottom
193: panel}) closely follows that in A5 for $r > R_2$, which is independent
194: of $\rho$ and $T$. As can be seen in Table~1, a reduction in $\rho$
195: owing to low $L_\nu$ in B5 (compared to that in A5) is accompanied
196: with a reduction in $T$. As a consequence, $\dot q_{\nu e}$ in B5 is
197: lower than that in A5 even for $r > R_2$. However, $\dot q_{\nu \nu}$
198: is not dependent on $T$ but is inversely proportional to $\rho$ (i.e.,
199: proportional to the number of neutrinos per volume), which becomes
200: significantly high owing to the decreasing $\rho$ for $r > R_2$
201: (Table~1). Note that the cooling terms ($\dot q_{e N} \propto T^6$ and
202: $\dot q_{e e} \propto T^9 \rho^{-1}$) quickly decay with increasing
203: $r$ and have negligible effects by the anisotropy (Fig.~2).
204:
205: As can be seen in the above numerical experiments, the strong
206: anisotropy in neutrino emission can be an additional energy source
207: pointed out by \citet{Qian96}. However, this mechanism may work only
208: for $r < 1.5\, R$, which is rather closer to the neutrino sphere than
209: the suggested range ($1.5 < r/R < 3$) by \citet{Qian96}. In the
210: current study, the effects of increasing entropy and accelerating wind
211: are prominent for the wind closer to the hot spot, in particular for
212: $R \approx 12\, \mathrm{km}$ (Table~1), at which $\dot q$ maximizes
213: (Fig.~2). The effect of anisotropic neutrino emission becomes less
214: important for a more distant wind (e.g., $R_2 = 20\, \mathrm{km}$),
215: where the neutrino heating has mostly ceased (Fig.~2). For a fixed
216: $R_2$, the effect is more significant for higher $L_{\nu 2}$ as can be
217: seen in Table~1. For model B5, the entropy is about $50 \%$ higher
218: ($180\, N_A\, k$) and the dynamic timescale is about a factor of ten
219: shorter ($1.65\, \mathrm{ms}$) than those in the isotropic model A1
220: with the same $L_\nu$ ($s = 117\, N_A\, k$ and $\tau_\mathrm{dyn} =
221: 14.1\, \mathrm{ms}$).
222:
223: Note that the purely parametric examinations explored in this section
224: should be regarded as only qualitative ones. For instance, the ``hot
225: spot'' is not necessary a point as illustrated in Figure~1. It is
226: conceivable that the area with strong neutrino emission has some
227: distribution on the neutrino sphere. Moreover, the configuration
228: should become multi-dimensional soon after the wind material passes
229: the point P$_2$ in Figure~1, which is treated within the framework of
230: a spherical wind model in the current study. More realistically, the
231: wind matter deviates from the radial to the direction of OP$_2$ in
232: Figure~1. This may moderate the acceleration of wind and increase the
233: heating duration. Hence, the current results may overestimate the
234: reduction of dynamic timescale and underestimate the increase of
235: entropy. It is difficult to estimate the net effect to the
236: nucleosynthesis from these modifications. Obviously, a
237: multi-dimensional approach will be needed to quantitatively estimate
238: the effects of the anisotropy.
239:
240: \section{Nucleosynthesis in Winds}
241:
242: Adopting the wind trajectories discussed in \S~2 for the physical
243: conditions, the nucleosynthetic yields are obtained by solving an
244: extensive nuclear reaction network. The network consists of 6300
245: species between the proton and neutron drip lines \citep[for more
246: detail, see][]{Wana06}. Neutrino-induced reactions and nuclear fission
247: are not considered in the current study. Each calculation is initiated
248: when the temperature decreases to $T_9 = 9$ (where $T_9 \equiv
249: T/10^9\, \mathrm{K}$). The initial compositions are given by $X_n = 1
250: - Y_{e}$ and $X_p = Y_{e}$, respectively, where $X_n$ and $X_p$ are
251: the mass fractions of neutrons and protons, and $Y_{e}$ is the initial
252: electron fraction (number of proton per nucleon) at $T_9 = 9$. In this
253: study, $Y_{e}$ is taken to be 0.4, according to the core-collapse
254: simulation in \citet[][at $L_\nu \approx 1 \times 10^{51}\,
255: \mathrm{ergs}\, \mathrm{s}^{-1}$]{Woos94}. As in \citet{Wana02}, the
256: temperature and density are set to be constant when $T_9$ decreases to
257: $1.0$, in order to mimic the effect of the slower outgoing ejecta
258: behind the shock.
259:
260: The nucleosynthesis results for models B2-B5, C2-C5, and D2-D5
261: (Table~1) are shown in Figure~3, as a function of atomic mass number.
262: For anisotropic wind models with $R_2 = 12\, \mathrm{km}$ (B2-B5), the
263: effect of anisotropic neutrino emission is evident. A factor of three
264: or four increase in $L_{\nu 2}$ (B3 and B4 in Table~1) leads to $s
265: \approx 150-160\, N_A\, k$ and $\tau_\mathrm{dyn} \approx 3-4\,
266: \mathrm{ms}$, resulting in the \textit{r}-process nucleosynthesis
267: (Fig.~3). For model B5, the high entropy ($= 180\, N_A\, k$) and short
268: dynamic timescale ($= 1.65\, \mathrm{ms}$) of the wind drive the
269: nuclear matter to the actinide region. The neutron-to-seed abundance
270: ratio at the beginning of the \textit{r}-process, defined as $T_9 =
271: 2.5$, is $Y_n/Y_h = 176$ and the final averaged mass number of heavy
272: nuclei with $Z > 2$ is $\langle A_h \rangle = 230$ (Table~1). For the
273: models with $R_2 = 15\, \mathrm{km}$ (C2-C5), the \textit{r}-process
274: still takes place when $L_{\nu 2}$ is four or five times higher than
275: $L_\nu$ (models C4 and C5). For $R_2 = 20\, \mathrm{km}$ (D2-D5), the
276: effect of anisotropic neutrino emission is not important and the
277: nucleosynthesis results are not significantly different from the
278: isotropic cases (A1-A5).
279:
280: \section{Implications}
281:
282: In this \textit{Letter}, the effects of anisotropy in neutrino
283: emission for the \textit{r}-process nucleosynthesis in
284: proto-neutron-star winds were examined, using the spherically
285: symmetric, steady outflow model of neutrino-driven winds. It was shown
286: that strong anisotropy, \textit{if it exists}, can be an additional
287: energy source \citep{Qian96} to heat the wind material. A factor of
288: four or five enhancement in \textit{effective} neutrino luminosity
289: results in the significant increase of entropy and shortening of
290: dynamic timescale of outgoing neutrino-heated ejecta. This is mainly
291: due to the \textit{boosted} neutrino heating from annihilation of
292: neutrino-antineutrino pairs into electron-positron pairs as a result
293: of anisotropic neutrino emission. This provides the physical condition
294: suitable for the robust \textit{r}-process, producing the third
295: abundance peak ($A = 195$) and beyond.
296:
297: It is conceivable that asymmetric neutrino emission can be associated
298: with the anisotropic matter distribution near the neutrino sphere. As
299: an example, \citet{Kota03} suggested that the non-spherical neutrino
300: sphere owing to rapid rotation leads to anisotropic neutrino heating
301: with the pole-to-equator ratio of a few to more than 10. This may
302: result in strong contrast in neutrino emission on the neutrino sphere,
303: which forms an \textit{effective} ``hot spot'' around the rotational
304: axis. A recent work with more sophisticated neutrino-transport scheme
305: by \citet{Wald05} showed, however, that the pole-to-equator flux ratio
306: is at most a factor of two, even for a rather rapidly rotating
307: core. This is a consequence that the radiation field is smoothened by
308: the many neutrino sources above the neutrino sphere (e.g., convective
309: bubbles) at the early phase ($< 1\, \textrm{s}$ after core
310: bounce). Nevertheless, all the convective bubbles are evacuated during
311: the late wind phase ($\sim 10\, \textrm{s}$) and a strong contrast of
312: neutrino flux might form on the neutrino sphere for a rapidly rotating
313: core.
314:
315: Another possibility of anisotropic neutrino emission might be due a
316: global fluid instabilities of neutrino-heated matter as observed in
317: multi-dimensional hydrodynamic simulations. Recent works have shown
318: that hydrodynamic instabilities can lead to low-mode ($l = 1$ in terms
319: of an expansion in spherical harmonics of order $l$) oscillation of
320: the convective fluid flow in the neutrino-heated layer behind the
321: shock \citep[e.g.,][]{Sche06, Bura06b, Burr06}. The presence of such a
322: low convective mode results in the pair of a single outflow and a
323: narrow accretion flow that creates the ``hot spot'' on the neutron
324: star surface. It should be noted, however, that the two-dimensional
325: simulations by \citet{Sche06} showed that the anisotropy of the
326: accretion luminosity owing to this flow appears to be only a few
327: percent (at least during the early phase up to $\sim 1\, \textrm{s}$
328: after core bounce). A future investigation relevant for the wind phase
329: ($\sim 1-10\, \textrm{s}$) will be needed to examine the degree of
330: anisotropic neutrino emission from such an accretion flow.
331:
332: Given one of the above (or another unknown) mechanism works, a
333: constraint for the \textit{r}-process may be obtained from the
334: condition that creates the ``hot spot'' owing to, e.g., rapid rotation
335: or long lasting accretion flow. It is conceivable that only a limited
336: fraction of supernovae create the ``host spot'' relevant for the
337: current scenario (e.g., rapid rotators or less-energetic supernovae
338: that form the long lasting accretion flow). This can be a reasonable
339: explanation for that the spectroscopic analysis of extremely
340: metal-poor stars and Galactic chemical evolution study imply only a
341: limited fraction of core-collapse supernovae undergo the
342: \textit{r}-process nucleosynthesis \citep{Ishi99, Ishi04}.
343:
344: The implications in this study must be tested by future
345: multi-dimensional simulations of core-collapse supernovae for long
346: duration ($\sim 10\, \mathrm{s}$) with accurate neutrino
347: transport. Systematic calculations of nucleosynthesis with such
348: hydrodynamic trajectories will be also needed to investigate the
349: contribution to the Galactic chemical evolution of \textit{r}-process
350: nuclei.
351:
352: \acknowledgements
353:
354: I would like to acknowledge H. -Th. Janka for helpful discussions. I
355: also acknowledge the contributions of an anonymous referee, which led
356: to clarification of a number of points in the original
357: manuscript. This work was supported in part by a Grant-in-Aid for
358: Scientific Research (17740108) from the Ministry of Education,
359: Culture, Sports, Science, and Technology of Japan.
360:
361: \begin{thebibliography}{}
362: \bibitem[Buras et al.(2006a)]{Bura06a}
363: Buras, R., Rampp, M., Janka, H. -Th, \& Kifonidis, K. 2006a, \aap,
364: 447, 1049
365: \bibitem[Buras et al.(2006b)]{Bura06b}
366: Buras, R., Janka, H. -Th., Rampp, M., Kifonidis, K. 2006b, \aap, submitted
367: (astro-ph/0512189)
368: \bibitem[Burrows et al.(2006)]{Burr06}
369: Burrows, A., Livne, E., Dessart, L., Ott, C. D., \& Murphy, J. 2006,
370: \apj, 640, 878
371: \bibitem[Ishimaru \& Wanajo(1999)]{Ishi99}
372: Ishimaru, Y. \& Wanajo, S. 1999, \apjl, 511, L33
373: \bibitem[Ishimaru et al.(2004)]{Ishi04} Ishimaru, Y., Wanajo, S., Aoki,
374: W., \& Ryan, S. G. 2004, \apj, 600, L47
375: \bibitem[Kotake et al.(2003)]{Kota03}
376: Kotake, K., Yamada, S., \& Sato, K. 2003, \apj, 595, 304
377: \bibitem[Otsuki et al.(2000)]{Otsu00}
378: Otsuki, K., Tagoshi, H., Kajino, T., \& Wanajo, S. 2000, \apj, 533, 424
379: \bibitem[Qian \& Woosley(1996)]{Qian96}
380: Qian, Y. -Z. \& Woosley, S. E. 1996, \apj, 471, 331
381: \bibitem[Scheck et al.(2006)]{Sche06}
382: Scheck, L., Kifonidis, K., Janka, H. -Th., M\"ueller, E. 2006, \aap,
383: submitted (astro-ph/0601302)
384: \bibitem[Sumiyoshi et al.(2000)]{Sumi00}
385: Sumiyoshi, K., Suzuki, H., Otsuki, K., Terasawa, M.,
386: \& Yamada, S. 2000, \pasj, 52, 601
387: \bibitem[Suzuki \& Nagataki(2005)]{Suzu05}
388: Suzuki, T. K. \& Nagataki, S. 2005, \apj, 628, 914
389: \bibitem[Thompson et al.(2001)]{Thom01}
390: Thompson, T. A., Burrows, A., \& Meyer, B. S. 2001, \apj, 562, 887
391: \bibitem[Thompson(2003)]{Thom03}
392: Thompson, T. A. 2003, \apjl, 585, L33
393: \bibitem[Walder et al.(2005)]{Wald05}
394: Walder, R., Burrows, A., Ott, C. D., Livne, E., Lichtenstadt, I.,
395: \& Jarrah, M. 2005, \apj, 626, 317
396: \bibitem[Wanajo et al.(2001)]{Wana01}
397: Wanajo, S., Kajino, T., Mathews, G. J., \& Otsuki, K. 2001, \apj, 554, 578
398: \bibitem[Wanajo et al.(2002)]{Wana02}
399: Wanajo, S., Itoh, N., Ishimaru, Y., Nozawa, S., \& Beers, T. C. 2002,
400: \apj, 577, 853
401: \bibitem[Wanajo(2006)]{Wana06}
402: Wanajo, S. 2006, \apj, 647, 1323
403: \bibitem[Woosley et al.(1994)]{Woos94}
404: Woosley, S. E., Wilson, J. R., Mathews, G. J., Hoffman, R. D., \&
405: Meyer, B. S. 1994, \apj, 433, 229
406: \end{thebibliography}
407:
408: %\clearpage
409:
410: \begin{deluxetable}{cccccccccc}
411: \tablecaption{Model Parameters and Results}
412: \tablewidth{0pt}
413: \tablehead{
414: \colhead{} &
415: \colhead{$R_2$} &
416: \colhead{$L_{\nu 2}$ (10$^{51}$} &
417: \colhead{$\dot M$ (10$^{-6}$} &
418: \colhead{$\rho_{13}$ (10$^7$} &
419: \colhead{$T_{13}$ (10$^{10}$} &
420: \colhead{$s$} &
421: \colhead{$\tau_\mathrm{dyn}$} &
422: \colhead{} &
423: \colhead{}
424: \\
425: \colhead{Model} &
426: \colhead{(km)} &
427: \colhead{ergs s$^{-1}$)} &
428: \colhead{$M_\odot\, \mathrm{s}^{-1})$} &
429: \colhead{g cm$^{-3}$)} &
430: \colhead{K)} &
431: \colhead{$(N_A\, k)$} &
432: \colhead{(ms)} &
433: \colhead{$Y_n/Y_h$} &
434: \colhead{$\langle A_h \rangle$}
435: }
436: \startdata
437: A1 & 10 & 1.0 & 3.86 & 3.49 & 2.06 & 117 & 14.1 & 6.01 & 107 \\
438: A2 & 10 & 2.0 & 13.3 & 5.98 & 2.35 & 103 & 7.20 & 8.16 & 109 \\
439: A3 & 10 & 3.0 & 27.4 & 8.10 & 2.54 & 95.5 & 4.99 & 6.02 & 107 \\
440: A4 & 10 & 4.0 & 46.0 & 9.75 & 2.66 & 90.7 & 3.90 & 6.02 & 107 \\
441: A5 & 10 & 5.0 & 68.7 & 11.9 & 2.80 & 87.0 & 3.26 & 10.3 & 111 \\
442: B2 & 12 & 2.0 & 4.88 & 2.68 & 1.88 & 131 & 7.45 & 18.0 & 118 \\
443: B3 & 12 & 3.0 & 5.84 & 2.08 & 1.73 & 145 & 4.44 & 38.9 & 135 \\
444: B4 & 12 & 4.0 & 6.84 & 1.65 & 1.61 & 161 & 2.72 & 80.7 & 170 \\
445: B5 & 12 & 5.0 & 7.93 & 1.15 & 1.46 & 180 & 1.65 & 176 & 230 \\
446: B5a\tablenotemark{a} & 12 & 5.0 & 5.84 & 2.07 & 1.74 & 147 & 4.37 & 39.8 & 136 \\
447: B5b\tablenotemark{b} & 12 & 5.0 & 6.89 & 1.64 & 1.60 & 162 & 2.61 & 86.9 & 174 \\
448: C2 & 15 & 2.0 & 4.15 & 3.37 & 1.99 & 127 & 9.76 & 12.1 & 113 \\
449: C3 & 15 & 3.0 & 4.41 & 3.25 & 1.94 & 136 & 7.08 & 20.9 & 120 \\
450: C4 & 15 & 4.0 & 4.68 & 3.25 & 1.90 & 147 & 5.15 & 34.7 & 131 \\
451: C5 & 15 & 5.0 & 4.97 & 3.10 & 1.84 & 159 & 3.69 & 58.3 & 153 \\
452: D2 & 20 & 2.0 & 3.93 & 3.45 & 2.04 & 122 & 11.9 & 8.43 & 109 \\
453: D3 & 20 & 3.0 & 4.00 & 3.53 & 2.04 & 127 & 10.2 & 11.3 & 112 \\
454: D4 & 20 & 4.0 & 4.08 & 3.50 & 2.02 & 132 & 8.69 & 14.9 & 115 \\
455: D5 & 20 & 5.0 & 4.15 & 3.48 & 2.01 & 137 & 7.43 & 19.3 & 119 \\
456: \enddata
457: \tablenotetext{a}{Same as B5, but without enhancement of $\dot q_{\nu \nu}$ for $r > R_2$.}
458: \tablenotetext{b}{Same as B5, but with enhancement of only $\dot q_{\nu \nu}$ for $r > R_2$.}
459: \end{deluxetable}
460:
461: %\clearpage
462:
463: \begin{figure}
464: \plotone{f1.eps}
465: \caption{Illustration of asymmetric neutrino emission. O is the center
466: of the neutron star. Strong neutrino emission from the ``hot spot''
467: near the point P$_1$ on the neutrino sphere is assumed, otherwise
468: being isotropic. A wind blowing from a nearby point P$_0$ with the
469: (weaker) isotropic neutrino emission ($L_\nu$) suddenly see a larger
470: number of neutrinos ($L_{\nu 2}$) when passing P$_2$.}
471: \end{figure}
472:
473: %\clearpage
474:
475: \begin{figure}
476: \plotone{f2.eps}
477: \caption{\textit{Top}: Net neutrino heating rates for wind models A1,
478: A5, B5, B5a, B5b, C5, and D5 listed in Table~1, as a function of
479: $r$. Jump of the heating rate at $r = R_2$ is due to the sudden
480: increase of effective neutrino luminosity from $L_\nu$ to $L_{\nu 2}$.
481: \textit{Bottom}: Heating ($\dot q_{\nu N}$, $\dot q_{\nu e}$, and
482: $\dot q_{\nu \nu}$) and cooling ($\dot q_{e N}$ and $\dot q_{e e}$)
483: rates as functions of $r$. Dashed and solid lines are for wind models
484: A5 and B5 (listed in Table~1), respectively.}
485: \end{figure}
486:
487: \clearpage
488:
489: \begin{figure}
490: \epsscale{2.0}
491: \plotone{f3.eps}
492: \caption{Final abundances obtained by the nucleosynthesis calculations
493: for wind models listed in Table~1 (except for A1-A5) as a function of
494: atomic mass number.}
495: \end{figure}
496:
497: \end{document}
498:
499: